factors influencing adolescent alcohol...
TRANSCRIPT
FACTORS INFLUENCING ADOLESCENT ALCOHOL AND MARIJUANA USE: THE
ROLE OF RELIGIOSITY, SCHOOL-BASED PREVENTION PROGRAMS, PARENTAL
INFLUENCE, AND PEER INFLUENCE
By
Ngoc N. Nguyen
M.S.W., Boston College, 2011
Submitted to the Graduate Faculty of
School of Social Work in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
University of Pittsburgh
2015
ii
UNIVERSITY OF PITTSBURGH
SCHOOL OF SOCIAL WORK
This dissertation was presented
by
Ngoc N. Nguyen
It was defended on
August 26th, 2015
and approved by
Gary K. Koeske, Ph.D., School of Social Work
Jeffery J. Shook, Ph.D., School of Social Work
Thomas Kelly, Ph.D., Western Psychiatric Institute and Clinic
Dissertation Committee Chairperson: Christina E. Newhill, Ph.D., School of Social Work
iii
Copyright © by Ngoc N. Nguyen
2015
iv
This study investigates the impact of personal and environmental factors, with greater
emphasis on the impact of religiosity on alcohol and marijuana use (Ys) among white, African
American, and Asian American adolescents. Specifically, this study aims to (1) examine if the
parental influence, peer influence, religiosity, and school-based prevention programs
independently and significantly predict the Ys, controlling for background factors; (2) explore
whether or not the expected impact of religiosity on Ys is qualified by race, gender and age; and
(3) explore if religiosity acts as a mediator of the relationships of age, race and gender with
alcohol and marijuana use. This study hypothesizes that (1) religiosity, school-based prevention
programs, parental support, parental monitoring, parental disapproval, peer use, and peer
disapproval will together significantly explain alcohol and marijuana use; and (2) higher
religiosity, attending alcohol and drug training programs, higher parental support, higher parental
monitoring, parental disapproval, peer disapproval, and less peer use will independently and
separately be related to lower likelihood of marijuana and alcohol use, controlling for
background factors.
The scope of this study aims at White, African American, and Asian American
adolescents aged 12 to 17 years old. A total of 12,984 adolescents were computed from the 2013
National Survey on Drug Use and Health (NSDUH) data. Separate binary logistic regression
analyses were conducted to examine the impact of individual religiosity, parental influence, peer
influence, and school-based prevention programs on alcohol and marijuana use among the study
FACTORS INFLUENCING ADOLESCENT ALCOHOL AND MARIJUANA USE:
THE ROLE OF RELIGIOSITY, SCHOOL-BASED PREVENTION PROGRAMS,
PARENTAL INFLUENCE, AND PEER INFLUENCE
Ngoc N. Nguyen, M.S.W
University of Pittsburgh, 2015
v
participants. Also, combination of OLS regression analysis and binary logistic regression
analyses was used to explore the moderation and mediation effects of religiosity, age, race, and
gender on alcohol and marijuana use among the study participants.
Findings confirm the study hypotheses. Results of exploratory analyses reveal that
religious girls are less likely to use alcohol and marijuana than religious boys; religiosity is not
impactful on alcohol and marijuana use among Asian American youth, which needs further
investigations; and religiosity can serve as a mediator on alcohol and marijuana use among
African American youth and female adolescents. Implications for social work practice, future
research, and drug policy are also discussed.
vi
TABLE OF CONTENTS
1.0 INTRODUCTION ........................................................................................................ 1
1.1 STATEMENT OF THE PROBLEM ................................................................. 1
1.2 ACOHOL AND DRUG PROBLEMS IN THE UNITED STATES ................ 2
1.2.1 Historical background .................................................................................... 2
1.2.2. Social values and beliefs ................................................................................. 4
1.2.3 Socio-economic status, race, and gender ....................................................... 5
1.2.4. Social stigmas and treatment services .......................................................... 7
1.2.5. Summary ......................................................................................................... 8
1.3 PURPOSE OF THE STUDY .............................................................................. 9
1.4 RESEARCH QUESTIONS ............................................................................... 10
1.5 HYPOTHESES .................................................................................................. 10
2.0 LITERATURE REVIEW .......................................................................................... 12
2.1 RELIGIOSITY AND ADOLESCENT SUBSTANCE USE .......................... 12
2.1.1 Type of substances ......................................................................................... 13
2.1.2 Communities .................................................................................................. 13
2.1.3 Religious measures ........................................................................................ 14
2.1.4 Race ................................................................................................................. 15
2.1.5 Age................................................................................................................... 16
2.1.6 Gender ............................................................................................................ 16
2.1.7 Summary and evaluation .............................................................................. 17
2.2 SCHOOL-BASED PREVENTION PROGRAMS ......................................... 18
2.3 PARENTAL INFLUENCE AND ADOLESCENT SUBSTANCE USE ....... 20
vii
2.3.1 Parental involvement ..................................................................................... 21
2.3.2 Parental support ............................................................................................ 22
2.3.3 Parental monitoring ...................................................................................... 23
2.3.4 Parental disapproval ..................................................................................... 24
2.3.5 Race ................................................................................................................. 25
2.3.6 Age................................................................................................................... 25
2.3.7 Gender ............................................................................................................ 26
2.3.8 Summary and evaluation .............................................................................. 26
2.4 PEER INFLUENCE AND ADOLESCENT SUBSTANCE USE .................. 27
2.4.1 Race ................................................................................................................. 30
2.4.2 Age................................................................................................................... 31
2.4.3 Gender ............................................................................................................ 31
2.4.4 Summary and evaluation .............................................................................. 31
3.0 THEORETICAL FRAMEWORKS ......................................................................... 33
3.1 SOCIAL LEARNING THEORY ..................................................................... 33
3.1.1 Key concepts and assumptions ..................................................................... 33
3.1.2 Analysis of conceptual frameworks ............................................................. 33
3.2 PROBLEM BEHAVIOR THEORY ................................................................ 34
3.2.1 Key concepts and assumptions ..................................................................... 34
3.2.1.1 The personality system ....................................................................... 34
3.2.1.2 The perceived environment system ................................................... 36
3.2.1.3 The behavior system ........................................................................... 37
3.2.2 Analysis of conceptual frameworks ............................................................. 38
viii
4.0 METHODOLOGY ..................................................................................................... 41
4.1 STUDY DESIGN AND PROCEDURE ........................................................... 41
4.2 PARTICIPANTS FOR THE CURRENT STUDY ......................................... 42
4.3 MEASURES ....................................................................................................... 42
4.3.1 Dependent variables ...................................................................................... 42
4.3.1.1 Marijuana use. ..................................................................................... 42
4.3.1.2 Alcohol use. .......................................................................................... 43
4.3.2 Predictors ....................................................................................................... 43
4.3.2.1 Religiosity. ............................................................................................ 43
4.3.2.2 School-based prevention programs. .................................................. 43
4.3.2.3 Parental support. ................................................................................. 43
4.3.2.4 Parental monitoring. ........................................................................... 44
4.3.2.5 Parental disapproval. .......................................................................... 44
4.3.2.6 Peer substance use. .............................................................................. 45
4.3.2.7 Peer disapproval. ................................................................................. 45
4.3.2.8 Race. ..................................................................................................... 46
4.3.2.9 Age. ....................................................................................................... 46
4.3.2.10 Gender. ............................................................................................... 46
4.4 DATA ANALYSES ............................................................................................ 46
5.0 RESULTS ................................................................................................................... 48
5.1 DESCRIPTIVE ANALYSES OF THE STUDY VARIABLES ..................... 48
5.2 BIVARIATE ANALYSES ................................................................................ 50
5.2.1 Bivariate analysis of all variables ................................................................. 50
ix
5.2.2 Bivariate analysis predicting marijuana use by all predictors .................. 50
5.2.3 Bivariate analysis predicting alcohol use by all predictors........................ 53
5.3 BINARY LOGISTIC REGRESSION ANALYSIS PREDICTING
MARIJUANA USE BY RELIGIOSITY, SCHOOL-BASED PREVENTION
PROGRAMS, PARENTAL INFLUENCE, PEER INFLUENCE, AND
DEMOGRAPHIC VARIABLES ...................................................................................... 55
5.4 BINARY LOGISTIC REGRESSION ANALYSIS PREDICTING
ALCOHOL USE BY RELIGIOSITY, SCHOOL-BASED PREVENTION
PROGRAMS, PARENTAL INFLUENCE, PEER INFLUENCE, AND
DEMOGRAPHIC VARIABLES ...................................................................................... 58
5.5 MODERATION AND MEDIATION TESTS ON THE IMPACTS OF
RACE, AGE, GENDER, AND RELIGIOSITY ON MARIJUANA AND ALCOHOL
USE. 60
5.5.1 Race, age, and gender as moderators for the relationships between
religiosity and alcohol and marijuana use ............................................................... 61
5.5.2 Religiosity as a mediator of race, age, and gender predicting marijuana
and alcohol use ........................................................................................................... 61
5.5.2.1 Tests of the Mediating Role of Religiosity in the Age, Gender and
Race effects on alcohol and marijuana use ...................................................... 62
6.0 DISCUSSION ............................................................................................................. 65
6.1 DISCUSSION OF SIGNIFICANT FINDINGS .............................................. 65
6.1.1 Findings of main analyses predicting marijuana and alcohol use ............ 65
6.1.2 Findings of moderation and mediation tests ............................................... 67
x
6.2 LIMITATION OF THE STUDY ..................................................................... 68
6.3 IMPLICATIONS ............................................................................................... 69
BIBLIOGRAPHY ....................................................................................................................... 73
xi
LIST OF TABLES
Table 1. Descriptive Analysis of Marijuana Use, Alcohol Use, Race, Gender, Parental
Monitoring, Parental Support, Parental Disapproval, Peer Disapproval, and School-Based
Prevention Programs Variables (N= 12,984) ................................................................................ 49
Table 2. Descriptive Analysis of Age, Number of Substance-Using Friends, and Religiosity
Variables (N=12,984) ................................................................................................................... 49
Table 3. Bivariate Analysis of all Predictors and Outcome Variables (N=12,984) ...................... 50
Table 4. Bivariate Analysis Predicting Marijuana Use by Race, Gender, Parental Monitoring,
Parental Support, Parental Disapproval, Peer Disapproval, and School-Based Prevention
Programs Variables (N=12,984) ................................................................................................... 52
Table 5. Bivariate Analysis Predicting Marijuana Use by Substance-Using Friends, Age, and
Religiosity (N=12,984) ................................................................................................................. 53
Table 6. Bivariate Analysis Predicting Alcohol Use by Race, Gender, Parental Monitoring,
Parental Support, Parental Disapproval, Peer Disapproval, and School-Based Prevention
Programs Variables (N=12,984) ................................................................................................... 54
Table 7. Bivariate Analysis Predicting Alcohol Use by Substance-Using Friends, Age, and
Religiosity (N=12,984) ................................................................................................................. 55
Table 8. Binary Logistic Regression Analysis Examining Marijuana Use (N=12,984) ............... 57
Table 9. Binary Logistic Regression Analysis Examining Alcohol Use (N=12,984) .................. 59
xii
LIST OF FIGURES
Figure 1. The conceptual structure of Problem Behavior Theory ................................................ 40
Figure 2. Gender as a moderator for the impact of religiosity on alcohol and marijuana use .... 60
Figure 3. Religiosity as a mediator for the impact of age, race, and gender on alcohol and
marijuana use................................................................................................................................ 60
Figure 4. Religiosity as a partial mediator for African Americans and females in marijuana use
(N=12,984) ................................................................................................................................... 64
Figure 5. Religiosity as a partial mediator for African Americans and females in alcohol use
(N=12,984) ................................................................................................................................... 64
1
1.0 INTRODUCTION
1.1 STATEMENT OF THE PROBLEM
Adolescent substance use is a big public health and public safety concern in the United States
(Humphreys & McLellan, 2010). According to the National Survey on Drug Use and Health
(NSDUH) in 2013, approximately 24.6 million Americans aged 12 or older were current illicit
drug users. Marijuana was the most commonly used illicit drug, accounting for 19.8 million
users or 80.6%. The report also revealed that current alcohol drinkers aged 12 or older were
136.9 million. Of this group, 16.5 million and 60.1 million people were heavy and binge drinkers
respectively. Alcohol and marijuana use increases with age. For youth aged 12 to 17, the rates of
marijuana use increased from 1.0% at ages 12 or 13 to 5.8% at ages 14 or 15 and to 14.2% at
ages 16 or 17. Similarly, the rate of alcohol use increased from 2.1% among persons aged 12 or
13 to 9.5% of persons aged 14 or 15, and to 22.7% of 16 or 17 year olds. Each year, substance
use costs the United States over $600 billion to cover expenses related to medical, economic,
criminal justice, and social impacts (SAMHSA, 2013).
Alcohol use is one of the main causes leading to morbidity and mortality among
adolescents (DHHS, 2007). Underage binge drinking is strongly correlated with other health
risks such as physical problems, unprotected sexual activity, physical and sexual assault, higher
risk for suicide and homicide, memory problems, changes in brain development, and even death
from alcohol poisoning (CDC, 2010; Miller, Naimi, Brewer, and Jones, 2007). Similar to teen
drinking, drug use is also attributable to negative health consequences such as cardiovascular
disease, stroke, cancer, HIV/AIDS, hepatitis B and C, lung disease, and mental disorders (NIDA,
2010). In addition to health problems, substance use also puts adolescents at high risks of poor
academic performance and increased school drop-outs (Chatterji, 2006; Malhotra & Biswas,
2
2006), increased peer substance use (Curran, Stice, and Chassin, 1997; Farrell and White, 1998),
and involvement in crime and violent activities (Corwyn & Benda, 2002; Popovici, Homer,
Fang, and French, 2012).
1.2 ACOHOL AND DRUG PROBLEMS IN THE UNITED STATES
Adolescent alcohol and marijuana use is part of alcohol and drug problems in the United States,
which is complex and unpredictable. Therefore, in order to have adequate knowledge of alcohol
and marijuana use among adolescents, it is necessary to understand socio-economic factors that
have a significant influence on the development of alcohol and drug problems in the country.
These factors may include history of the United States, social values and beliefs, demographics,
as well as social stigmas and inadequate treatment.
1.2.1 Historical background
Alcohol and alcoholism, which are critical parts of many Americans’ lives, have a very long
history (Kleiman & Hawdon, 2011). Over the past centuries, both the colonists and the U.S
government have tried to ban alcohol and control alcoholism several times such as promulgation
of the 1672 law prohibiting the payment of wages in alcohol, and the Volstead Act or the
National Prohibition Act in 1919 outlawing the manufacture, importation, exportation, sale,
distribution, and transportation of alcohol. However, the efforts were not successful for many
reasons such as increasing smuggling alcohol at large scale, illicit manufacture of alcohol, and
costs related to law enforcements (Korsmeyer & Kranzler, 2009). Finally the U.S government
officially legalized alcohol content of 3.2% after the ratification of the Twenty-first Amendment
in 1933 followed by the Cullen-Harrison Act (Levinson, 2000; Morgan, 1981).
Similar to alcohol, other illicit drugs such as opium, morphine, marijuana, morphine,
cocaine, and amphetamine have been used for both recreational and medical purposes in The
3
U.S. for many years. For example, marijuana was commonly used for recreation and medication
as an anticonvulsant and relaxant among the Mexican immigrants during the mid-nineteenth
century. By early 20th century, the drug problems drastically increased among men, women, and
children in the U.S. There were several reasons resulting in the drug problems such as the
returning home of addicted American soldiers from World War I, the influx of illicit drugs to the
U.S. smuggled by organized criminal gangs, and rebelling of many young baby boomers – the
hippie subculture (Gahlinger, 2001; Durrant & Thakker, 2003; Korsmeyer & Kranzler, 2009). As
drug use and addiction were blamed for the causality of social evil and crime, the U.S.
government gradually took action to end the laissez-faire approach to drugs and control the
substances through promulgation of numerous laws such as the Pure Food and Drug Act of 1906,
the Harrison Act of 1914, the Narcotic Drug Import and Export Act or Jones-Miller Act, the
Heroin Act of 1924, and the Marijuana Tax Act of 1937 (Kleiman & Hawdon, 2011; Durrant &
Thakker, 2003). For example, the Harrison Act of 1914 regulated the use of opiates and cocaine
for non-medical purposes as an illegal behavior, which transformed drug addicts from patients to
criminals (Acker, 1993). The U.S. government’s view toward addiction treatment fluctuates over
time. Treatment clinics were first established in 1913, then were shut down in 1925, and were
not re-opened until another decade with the initiation of two prison-liked narcotic farms - at
Lexington, Kentucky in 1935 and in Fort Worth, Texas in 1938 (White, 1998). Addiction was
not treated as a disease instead of crime until after a declaration of the Supreme Court in 1962
(Levinson, 2002). Recently, Office of the National Drug Control Policy (ONDCP) has released
the 2014 National Drug Policy Strategy which clearly states that addiction is a brain disease that
can be prevented, treated for recovery, and not a moral failure on the individual. According to
the new strategy, the U.S. government will implement comprehensive measures such as
4
increasing preventive methods, providing early intervention, making access to treatment,
eliminating barriers to recovery, and reforming the criminal and juvenile justice system which
inclines towards treatment versus incarceration for non-violent and low-level offenders
(ONDCP, 2014).
Adolescent drug problems were at crisis level in the 1960s. The U.S. government and
nonprofit organizations made great efforts to prevent adolescents from using drugs. Nonprofit
organizations such as PRIDE (Parents’ Resource Institute for Drug Education) took a lead in a
strong movement against drug use, especially marijuana among school students in the U.S. in the
1970s and 1980s (Levinson, 2002; Durrant & Thakker, 2003). They brought parents together to
share drug information and protect their communities from drug influence. Similarly, the U.S
government also implemented various school-based drug prevention programs such as D.A.R.E
(Drug Abuse Resistance Education) to deal with adolescent drug problems in 1983(Korsmeyer &
Kranzler, 2009). However, these programs were ineffective and even exacerbated adolescent
drug problems (Braucht, Follingstad, Brakarsh, and Berry, 1973; Randall & Wong, 1976).
Adolescent substance use is still increasing and unsolvable in the U.S as seen in the 2013
NSDUH report.
1.2.2. Social values and beliefs
Alcohol and drug problems are complex and they are viewed differently over time in the U.S.
During the laissez-faire period (prior to the 1906 Pure Food and Drug Act) alcohol and drugs
were freely sold in the market for any purposes. At that time, nobody including physicians
regarded alcohol and drugs as social problems (Durrant & Thakker, 2003; Gahlinger, 2001;
White, 1998). However, social attitudes and beliefs toward substance use gradually changed due
to addiction and drug-related problems such robbery and other criminal activities (Korsmeyer &
5
Kranzler, 2009; Levinson, 2002). Additionally, there were other latent reasons contributing to
social concerns about substance use and addiction. For example, Whites framed opium smoking
and marijuana use, which are part of custom of the Chinese and Mexican immigrants, as an
immoral sign and social stigma in order to compete with these low-paid workforces (Korsmeyer
& Kranzler, 2009).
There are different beliefs about the causality of addiction. Many Americans blame
addiction on the development of machine-age life, low moral standards, over prescribing
medications subject to abuse, and inadequate law enforcement (Levinson, 2002). Meanwhile, the
U.S. government believes that addiction is a consequence of both supply and demand sides.
Internationally they collaborate with other foreign countries such as Mexico and Columbia to
reduce drug supply to the U.S. Domestically they attack the demand through law enforcement
and treatment services (Durrant & Thakker, 2003; Morgan, 1981). However, the “war on drugs”
drug policy is not effective as it fails to eliminate the drug problem in America.
1.2.3 Socio-economic status, race, and gender
The correlation between socioeconomic status and substance use is quite complex and varies
significantly among studies. Goodman and Huang (2002), in a cross-sectional study, found that
adolescents living in low SES families, as measured by household income and parental
education, were more vulnerable to alcohol and cocaine use than those who lived in affluent
families. However, there is also evidence that adolescents with higher SES have greater risks for
developing substance use behaviors. Three cross-sectional studies showed that adolescents
growing up in higher SES families were more likely to use substances than those who were born
in lower SES families (Blum et al., 2000; Humensky, 2010; Hanson & Chen, 2007). For high
SES adolescents, family income is a stronger predictor of substance use than family status
6
(Hanson & Chen, 2007). According to the researchers, it may be that the availability of financial
resources is more influential on teen substance use than the social status associated with having
parents with high education and good jobs. Neighborhood SES is also predictive of adolescent
substance use, and this correlation is moderated by parental substance use. Trim & Chassin
(2008), in a longitudinal study, found that children of non-alcoholics were at higher risk of
alcohol use, living in a higher SES neighborhood; and children of alcoholics were more
susceptible to higher risk of alcohol use, living in lower SES neighborhood.
Adolescent substance use affects across all races. Still, its impact is different from race to
race. Four cross-sectional studies showed that White adolescents had higher rates of substance
use than Black, Hispanic, and Asian Americans (Blum et al, 2000; Mason, Mennis, Linker,
Bares, & Zaharakis, 2013; Thai, Connell, and Tebes, 2010; Barnes, Welte, and Hoffman, 2002).
The finding is supported by Tanner-Smith (2012) who conducted a longitudinal study and found
that White adolescents had the highest level of alcohol and marijuana use at follow-ups in
comparison with Hispanic and Black. Asian adolescents are reported to have the lowest level of
alcohol use, binge drinking, and illicit drug use in comparison with Whites, Blacks, Hispanics,
West Indians, American Indians, and other races in the U.S. (Barnes, Welte, and Hoffman,
2002). Meanwhile, American Indian youth have the highest levels of alcohol use, binge drinking,
and illicit drug use (Barnes, Welte, and Hoffman, 2002). Cultural, socialization, and individual
factors could be predictive of the racial differences in the study, with such factors protecting
Asian youth and putting American Indian adolescents at higher risk of substance use (Barnes,
Welte, and Hoffman, 2002).
Adolescent substance use is also different among males versus females. Cross-sectional
studies show that males are more sensitive and susceptible to substance use than females
7
(Barnes, Welte, and Hoffman, 2002; Svensson, 2003). There is also a difference among females
and males related to racial differences. White females drink more alcohol, and black males use
more marijuana than other ethnic groups (Mason, Mennis, Linker, Bares, & Zaharakis, 2013).
Lack of parental supervision is predictive of the development of adolescent substance use. Both
male and female adolescents who use drugs often have less parental supervision than those who
do not use drugs, regardless of SES (Svensson, 2003). Additionally, peer attitude significantly
contributes to both male and female substance use (Mason, Mennis, Linker, Bares, & Zaharakis,
2013).
1.2.4. Social stigmas and treatment services
Statistics show that not many individuals with alcohol and drug problems have access to
treatment services due to inadequate availability of treatment programs (Lo & Cheng, 2011;
SAMHSA, 2012). Treatment services for adolescents are both inadequate and underdeveloped;
they largely depend on models for adult treatment which do not take into account adolescents’
developmental stages (Cavanaugh & White, 2003). Currently, there is a lack of empirically
supported outpatient treatment programs which specifically meet the needs of adolescents with
alcohol and marijuana problems (McWhirter, 2008). Also, there is a little attention given to the
practice settings, service delivery systems, and staff’s qualification (Cavanaugh, Kraft, Muck, &
Merrigan, 2011). There are numerous reasons leading to inadequate treatment for adolescents
such as lack of coordination among federal and state agencies, differences between federal and
state agencies in using resources, fragmentation of child serving services, inadequate service
delivery, and lack of qualified staff (Cavanaugh & White, 2003). Over the past years, many
evidence-based and behavioral treatment programs for adolescents such as Motivational
Interviewing, Multidimensional Family Therapy, and 12 step programs (NIDA, 2012) have been
8
implemented to provide services for adolescents who use substances. However, the effects of
these programs are not always confirmed. For example, Barnett and colleagues (2012) conducted
a meta-analysis of 39 Motivational Interviewing studies on adolescent drug use, including two
quasi-experimental studies and 37 randomized control trials (31 randomized by individuals and 6
randomized by groups) in various settings. They found that 28 of the 39 studies (72%) showed
significant reductions in drug use, including seven studies on alcohol use, six studies on tobacco
use, seven studies on marijuana use, and eight studies on other drug use. Eleven studies including
four on tobacco, two on alcohol, two on marijuana, and three on other drugs showed no effect at
all.
Social stigmas also create barriers for people who have substance abuse problems to
access to treatment (McFarling et al, 2011). A report by Clinical Practice Guideline Treating
Alcohol and Drug Use and Dependence (2008) revealed that drug users were often described
with such words as “sinner”, “irresponsible”, “selfish, and “weak”. Such stigmas make drug
users fear and prevent them from seeking help (Erickson, 2007) and reflects the long-standing
“moral model” of addiction etiology.
1.2.5. Summary
Alcohol and drug problems in the U.S. are consequences of numerous structured elements such
as historical legacy, socio-economic condition, social values and beliefs, political perspectives,
inadequate treatment, and social stigma. These elements either contribute to the development of
the problems or hinder efforts to solve the problems (McFarling et al, 2011). So far, substance
use has expanded to all races, classes, ages, and gender in America. Adolescents are the most
vulnerable population as they are more likely to get involved in substance use due to
9
environmental and developmental factors. Despite numerous efforts, adolescent substance use
problem is still unsolvable and increasing among adolescents.
1.3 PURPOSE OF THE STUDY
Given the increasing alcohol and marijuana problems, complex history of the problems in the
United States, and negative consequences of the problems, this study aims to explore factors
influencing the use of substances among adolescents. As living in a social context, the initiation
and development of substance use among adolescents are strongly influenced by personal and
environmental factors such as religious beliefs, family, and friends, which are inter-relatedly
connected. Besides, school-based prevention programs, which provide adolescents with
knowledge of substance use and coping skills, play an important role in deterring or decreasing
substance use among this population. Examining the effects of personal and environmental
factors, and prevention programs is not a new area of research. However, none of studies in the
past have examined the effects of all of these factors together in one study. In addition, it is
worth using a large national sample from a most recent data set to re-examine the effects of these
factors with greater emphasis on individual religiosity. Therefore, this study aims to:
1. report descriptive statistics on independent and dependent variables;
2. evaluate relationships between a set of anticipated predictors of marijuana and alcohol
use and report the aggregate amount of explanation they provide;
3. examine if the predictors independently and significantly predict the Ys, controlling for
basic background factors, and focusing in particular on the influence of religiosity;
4. explore whether or not the expected impact of religiosity on Y is qualified by race,
gender and age. In other words, is the effect invariant across these major background variables?
10
5. explore if religiosity acts as a mediator of the expected relationships of age, race and
gender with alcohol and marijuana use.
1.4 RESEARCH QUESTIONS
1. (A) Do individual religiosity, school-based prevention programs, parental influence, and
peer influence significantly predict adolescent alcohol and marijuana use as have been found in
previous research?; (B) Do they remain predictors controlling for background (age, race, and
gender) in a multivariate context?
2. Is the anticipated influence of religion on lower alcohol and marijuana use moderated by
age, gender and race of the youth? Some past research suggests that females, African-Americans
and younger youth may be particularly less likely to use drugs and alcohol if they are more
religious.
3. Does religiosity act as a mediator of the presumed tendency for younger, African
American and female youth to use marijuana and alcohol with lower likelihood? The rationale
for this question is that religious youth (Z) have been shown to be less likely to use alcohol and
marijuana (Y). And, the background variables (Xs) of age, gender and race have been found to
relate to amount of religiosity. Analyses will be done to determine if the obtained relationships
are consistent with mediation and thereby provide a basis for more formal mediation analyses
subsequently.
1.5 HYPOTHESES
1. Religiosity, school-based prevention programs, parental support, parental monitoring,
parental disapproval, peer use, and peer disapproval will together significantly explain alcohol
and marijuana use.
11
2. Higher religiosity, attending alcohol and drug training programs, higher parental support,
higher parental monitoring, parental disapproval, peer disapproval, and less peer use will
independently and separately be related to lower likelihood of marijuana and alcohol use,
controlling for background factors.
12
2.0 LITERATURE REVIEW
2.1 RELIGIOSITY AND ADOLESCENT SUBSTANCE USE
The extant literature on religiosity has focused on two main areas including individual-level
religiosity and community-level religiosity. Individual-level religiosity is often measured by six
dimensions including (1) Church attendance; (2) Salience (the influence and importance of
religiosity); (3) Denomination affiliation (e.g., Catholic, Baptist, Jewish, etc.); (4) Prayer; (5)
Bible study; and (6) Religious activities both inside and outside of typical church settings
(Johnson, De Li, Larson, and McCullough, 2000). Community-level religiosity is measured by
the church membership of the individuals in that community (Regnerus, 2003; Wallace et al.,
2007). The effects of individual religiosity on adolescent substance use are inconsistent among
the extant research. Many have confirmed that individual religiosity has an inverse or negative
relationship with adolescent substance use (Sloane & Potvin, 1986; Stark, 1996; Wallace,
Brown, Bachman, & LaVeist, 2003; Wallace et al., 2007; Vaughan, de Dios, Steinfeldt, & Kratz,
2011; Bahr & Hoffmann, 2008). However, others have argued that there are no deterrent effects
of individual religiosity on adolescent substance use (Bahr, Hawks, & Wang, 1993; Marcos,
Bahr, & Johnson, 1986). To clarify this controversy, Johnson, De Li, Larson, and McCullough
(2000) conducted a systematic review of 40 studies on the effects of religiosity and revealed that
86% of the studies reported negative effects, or religiosity decreased substance use; One study
found positive effect, or religiosity increased substance use; and the remaining studies found
either non-significant or inconclusive effects. The relationship between individual religiosity and
adolescent substance use depends on numerous factors such as type of substances, communities
that adolescents belong to, religious measures, race, gender, and age of adolescents.
13
2.1.1 Type of substances
The effects of individual religiosity on adolescent substance use vary significantly and depend on
type of substances. Individual religiosity has more deterrent effects on alcohol and marijuana use
than other hard illicit drugs such as cocaine, heroin, and amphetamine. Most of the existing
studies have confirmed that individual religiosity increases abstinence and decreases alcohol and
marijuana use among adolescents (Jang & Johnson, 2001; Vaughan, de Dios, Steinfeldt, and
Kratz, 2011; Johnson, Larson, and McCullough, 2000; Stark, 1996; Kelly, Pagano, Stout, and
Johnson, 2011). Only a few studies have contended that there is no deterrent effect of individual
religiosity on the use of marijuana and alcohol among youth (Bahr, Hawks, & Wang, 1993;
Marcos, Bahr, & Johnson, 1986). For other hard illicit drugs, only one study has confirmed that
individual religiosity has a negative effect on hard illicit drug use, or religious youth are less
likely to use hard illicit drugs (Jang & Johnson, 2001). Meanwhile, more others have confirmed
that individual religiosity fails to prevent adolescents from using hard illicit drugs (Bahr &
Hoffmann, 2008; Bahr, Hawks, & Wang, 1993; Marcos, Bahr, & Johnson, 1986).
2.1.2 Communities
Communities refer to moral or secular communities (regions, schools or neighborhoods) that
adolescents belong to. The extant literature reveals inconsistent findings about the effects of
individual religiosity on adolescent substance use in moral sectarian (those with high rates of
religious participation) and secular community (those with low rates of religious participation).
Some studies have concluded that frequency of church attendance and the importance of religion
are protective factors to decrease substance use for adolescents living in religious communities,
but not for those who live in secular communities (Stark, 1996; Wallace et al., 2007; Baier &
Wright, 2001). In contrast, Tittle & Welch (1983) argued that individual religiosity as measured
14
by frequency of church attendance had greater impact on marijuana use in secular community
than in moral community. According to Tittle & Welch (1983), religious adolescents in a secular
community are less likely to use marijuana than religious counterparts who live in a religious
community. Meanwhile, there is also a neutral trend that individual religiosity has equal effects
on adolescent substance use in both of religious and secular communities, in other words, there is
no significant difference in the effects of individual religiosity on substance use in these two
communities (Chadwick & Top, 1993).
The impact of individual religiosity on substance use is not similar among religious
communities. Stark (1996) found that individual religiosity (church attendance) had a strong
negative correlation with alcohol use among Protestants, but it had no impact on Catholics.
According to this finding, Protestants who frequently attend church are less likely to drink
alcohol; however, frequently attending church does not prevent Catholics from using alcohol.
This study also addressed that individual religiosity protects both Protestants and Catholics from
using marijuana, but the strength of protection is a bit weak among Catholics (Stark, 1996).
2.1.3 Religious measures
As reviewed by Johnson, De Li, Larson, and McCullough (2000), the two most commonly used
religious measures in the existing studies are church attendance and salience. The effects of these
religious measures are inconsistent among studies. Some studies have found that church
attendance and the importance of religiosity have no deterrent effects on adolescent substance
use (Bahr, Hawks, & Wang, 1993; Marcos, Bahr, & Johnson, 1986). Conversely, many others
have confirmed that these religious measures can deter adolescents from using alcohol and
marijuana (Sloane & Potvin, 1986; Jang & Johnson, 2001; Stark, 1996; Hirschi & Stark, 1969;
Wallace et al., 2007; Wallace et al., 2007; Wallace, Brown, Bachman & LaVeist, 2003; Rote &
15
Starks, 2010; Bahr & Hoffmann, 2008; Vaughan, de Dios, Steinfeldt & Kratz, 2011). Comparing
the strength of church attendance and salience, the current literature reveals inconclusive
findings. Two studies found that salience as indicated by influence of religiosity and the
importance of religiosity had a stronger effect than church attendance (Sloane & Potvin, 1986;
Regnerus & Elder, 2003). Inversely, Rote & Starks (2010) argued that church attendance had
larger deterrent effects than the importance of religion. However, combination of church
attendance and the importance of religiosity are effective to decrease alcohol use and increase
abstinence among adolescents (Regnerus & Elder, 2003). Unlike church attendance and salience,
denominational affiliation is not as effective as these two measures. Two studies have concluded
that denominational affiliation had relatively small deterrent effects on adolescent substance use
(Wallace et al., 2007; Wallace, Brown, Bachman, & LaVeist, 2003). Among religious
denominations, Protestants are less likely to drink than Catholics as historically Protestant
doctrine strongly opposes drinking (Stark, 1996).
2.1.4 Race
Studies on racial differences in substance use have found that Black youth are more religious
than White youth; Black youth are more likely than White youth to abstain from using
substances; the strength of the inverse relationship between religiosity and substance use is
stronger for White youth than Black youth; and Hispanic youth are in between White and
African American youth in terms of abstention from substance use (Brown, Parks, Zimmerman,
& Phillips, 2001; Rote & Starks, 2010; Wallace, Brown, Bachman, & LaVeist, 2003).
Religious measures make a significant contribution to the racial effects of religiosity.
Studies have concluded that church attendance and the importance of religion have equal
deterrent effects on substance use for White, Black, and Hispanic adolescents (Wallace et al.,
16
2007; Vaughan, de Dios, Steinfeldt, & Kratz, 2011; Rote & Starks, 2010). However, findings are
not similar to denominational affiliation. Wallace, Brown, Bachman, & LaVeist (2003) found
that Black adolescents were more likely than White counterparts who are affiliated with similar
denominations to abstain from alcohol use. So far, there is a dearth of studies on the impact of
religiosity on substance use among Asian American adolescents. One study found that individual
religiosity had no deterrent effect on substance use behavior of Asian Americans (Chung, 1997).
2.1.5 Age
The effects of religiosity on substance use vary across ages, depending on type of substances.
Jang and Johnson (2001) concluded that the effects of individual religiosity on hard illicit drug
use increased with the development of adolescents; however, this is not the case for marijuana.
The researchers found that the religious effects on marijuana use were stronger between early
and later adolescence, peaked at ages of later adolescence, and then slowly declined thereafter
(Jang and Johnson, 2001). The effects of religiosity on alcohol use have not been reported so far.
2.1.6 Gender
Numerous studies have confirmed that girls are more religious than boys (Hoffmann & Johnson,
1998; Donahue & Benson, 1995; Hood, Spilka, Hunsberger, & Gorsuch, 1996; Wallace, Forman,
Caldwell, and Willis, 2003; Salas-Wright, Vaughn, Hodge, & Perron, 2012). Regarding the
strength of the impact of religiosity on adolescent substance use, studies have concluded that
religiosity is stronger among girls than among boys, indicating that religious girls are less likely
to use or more likely to abstain from using alcohol and marijuana than religious boys (Wills,
Yaeger, & Sandy, 2003; Pitel et al., 2012)
17
2.1.7 Summary and evaluation
In general, religiosity has more deterrent effect on alcohol and marijuana use than other hard
illicit drugs. The effects of individual religiosity on adolescent substance use vary a lot in
different communities, depending on numerous factors such as type of religions, religious
measures, and type of substances. The existing studies mainly focus on White, Black, and
Hispanic adolescents, whereas there is a dearth of studies on Asian American population. The
impact of religiosity on adolescent substance use also depends on age, race, gender, and type of
substances.
The inconsistent findings among studies could be explained by three main factors
including methodological limitations, dimension of religious measures, and sampling. Regarding
methodological limitations, the majority of the existing studies were based on cross-sectional
data, and half of them did not test the reliability of the religious measures (Johnson, De Li,
Larson, and McCullough, 2000). Additionally, validity of the previous findings is of concern.
Ordinary Least Squares (OLS) regression - the most commonly-used statistical technique for
data analysis erroneously assumes that students’ responses are independent and it does not take
into account the school context. In fact, 85% of the current studies were drawn from school
students whose behaviors are significantly influenced by the school context (norms, culture, and
social environments). Therefore, the relationship between individual religiosity and substance
use could have been erroneously interpreted (Baier & Wright, 2001). Dimension of religious
measures is another issue. Johnson, De Li, Larson, and McCullough (2000) confirmed that only
studies that used four or more religious measures consistently yielded negative or beneficial
effects of individual religiosity on substance use; studies that used three or less dimensions
reported mixed or inconclusive findings. Meanwhile, 60% the existing studies used only one or
18
two dimensions of religious measures (Johnson, De Li, Larson, and McCullough, 2000). With
regards to sampling, Baier & Wright (2001) concluded that religiosity had stronger deterrent
effects on adolescent substance use in studies using small sample sizes, more racially diverse
samples, and data collected later in time.
These findings shed light on critical information for future studies. When examining the
relationship between religiosity and adolescent substance use, it is essential to test reliability of
religious measures and use statistical techniques that can ensure validity of findings.
Additionally, it is strongly encouraged to use four or more dimensions of religious measures,
recently collected data, and racially diverse samples for accuracy of research findings.
2.2 SCHOOL-BASED PREVENTION PROGRAMS
Since the outbreak of adolescent substance use problems in 1960s, numerous school-based
prevention programs have been implemented in the United States. However, the effectiveness of
these school-based programs is still inconclusive. Two longitudinal studies evaluating the
effectiveness of school-based drug prevention programs concluded that there were no significant
differences between treatment and control groups, or between pretest and posttest results,
indicating that these programs failed to have positive impact on adolescents’ knowledge,
attitudes, and the use of substances (Webster, Hunter, & Keats, 2002; Bonaguro, Rhonehouse, &
Bonaguro, 1988). Whereas, others found that school-based prevention programs increased
knowledge, attitudes, and interpersonal skills, and decreased the use of tobacco and marijuana,
but not alcohol use among adolescents (Hansen, Malotte, & Fielding, 1988; Botvin, Baker,
Dusenbury, Tortu, & Botvin, 1990). The effectiveness of school-based prevention programs
largely depends on types of program. In a meta-analysis of 207 universal school-based drug
prevention programs, Tobler et al. (2000) concluded that interactive programs, which foster the
19
development of interpersonal skills (refusal, communication, assertive, decision-making, and
coping skills), had stronger effects than non-interactive programs, which focus on drug
knowledge and affective development (self-esteem, self-awareness, attitudes, beliefs, and
values).
Created by Los Angeles police Chief Darryl Gates in 1983, Drug Abuse Resistance
Education (D.A.R.E) is a widely-known drug prevention program for school-age students in the
United States. The program was initially designed to educate fifth and sixth graders about drugs
and provide them with decision making skills to say no to drugs. Presently, the program is
expanded to older students. D.A.R.E training curriculum focused on (1) understanding the effects
and consequences of drug use; (2) recognizing and coping with interpersonal pressures to drug
use; (3) promoting self-esteem and assertiveness; (4) providing positive alternatives; and (5)
increasing students’ interpersonal communication and decision-making skills. D.A.R.E lectures
are given by uniformed police officers who have undergone an intensive 80-hour training course
on various skills such as public speaking, teaching methods, and classroom management in
addition to the core curriculum. Although D.A.R.E is widely applied to 80% of schools in the
United States and 40 other countries (Mahon-Halt & Mosher, 2011) the effectiveness of this
program is still controversial. Ennett, Tobler, Ringwalt, & Flewelling (1994), who conducted
meta-analysis review of eight D.A.R.E evaluations, concluded that short-term effects of D.A.R.E
on reducing or preventing drug use behavior was small; and the program was even less effective
than other interactive prevention programs. In another review of D.A.R.E outcomes, Dukes,
Ullman, & Stein (1996) revealed that immediately after completion, D.A.R.E increased self-
esteem and institutional bonds (with family, police, and teachers) and decreased risky behaviors,
but there were no significant differences between D.A.R.E. and comparison groups. Whereas
20
Dukes, Ullman, & Stein (1995) found that participants in two D.A.R.E groups had greater self-
esteem, stronger institutional bonds, and fewer risky behaviors than participants in two control
groups when controlling for maturation effects. Critics addressed several reasons which led to
the ineffectiveness of D.A.R.E such as questionable delivery methods by the police, lack of
scientific knowledge on the effects of drugs, exaggeration of risks, strong prohibition of risk, and
lack of social support (Mahon-Halt & Mosher, 2011). Since the year of 2000, training curriculum
and teaching methods of D.A.R.E have been improved over time. The most recent D.A.R.E
curriculum, which was revised in 2009, is D.A.R.E REAL (Refuse, Explain, Avoid, and Leave).
The new curriculum is purportedly based on scientific findings about drugs, cognitive behavioral
therapy, and motivational enhancement therapy techniques (Mahon-Halt & Mosher, 2011).
However, the long-term effectiveness of this program is still pending.
2.3 PARENTAL INFLUENCE AND ADOLESCENT SUBSTANCE USE
Hirschi (1969) posited that social and cultural constrains, which are strongly associated with
parental influence, are critical factors that prevent adolescents from committing acts of deviance.
Inept parenting leads to socially unskilled adolescents, who are consequently more likely to join
deviant peer groups in which substance use occurs (Hawkins & Weis, 1985; Patterson,
DeBaryshe, & Ramsey, 1989). Conversely, conventional bonds such as parental involvement and
monitoring can deter or lower levels of substance use during adolescence (Erickson, Crosnoe, &
Dornbusch, 2000). The scope of this section will focus on parental factors that may prevent
adolescents from using substances or reduce their problems, including parental involvement,
parental support, parental monitoring, and parental disapproval. Other demographic factors such
as age, gender, and race will be also discussed in the relationship with parental factors.
21
2.3.1 Parental involvement
This is a broad term and it is categorized by different components such as shared
communication, shared activities, and emotional closeness in some studies. In general, parental
involvement has an inverse relationship with adolescent substance use, indicating that parental
involvement deters or prevents adolescents from using alcohol and marijuana (Wills, Resko,
Ainette, & Mendoza, 2004; Barnes, Reifman, Farrell, & Dintcheff, 2000; Whitney, Kelly, Myers,
and Brown, 2002; Pilgrim, Schulenberg, O'Malley, Bachman, & Johnston, 2006). For example,
in their study, in which parental involvement was measured by helping adolescents do
homework, requiring them to do chores, and setting limit for TV watching, Pilgrim, Schulenberg,
O'Malley, Bachman, & Johnston (2006) found that parental involvement had both direct and
indirect impact on adolescent alcohol and marijuana use. Specifically, parental involvement
significantly decreased the frequency of alcohol and marijuana use among both 8th and 10th
graders in the study, and this relationship was mediated by school success and time with friends
(Schulenberg, O'Malley, Bachman, & Johnston, 2006). However, parental over-involvement and
control can be as risk factors for excessive alcohol use among adolescents (Dishion & Loeber,
1985).
With regard to the components of parental involvement, their effects on adolescent
substance use vary significantly. Cross-sectional and longitudinal studies found that adolescents
who feel close to their parents were less likely to drink alcohol or get drunk than those who do
not; and shared activities such as sports, religious services, social outings, shopping, and school
projects with their parents can protect adolescents from using substances (Goncy & Van Dulmen,
2010; Lewis & Jordan, 2005). Conversely, these researchers also revealed that shared
communication was positively associated with adolescent substance use, indicating that the
22
greater shared communication with their parents the more likely adolescents drink alcohol or use
marijuana (Goncy & Van Dulmen, 2010; Lewis & Jordan, 2005). One possible explanation of
this counter-intuitive relationship is that the shared communication could have invoked the
child’s negative feelings such as hostility or wariness, which consequently promote their
substance use (Pleck & Masciadrelli, 2004).
2.3.2 Parental support
Similar to parental involvement, studies have confirmed that parental support has a direct and
inverse relationship with adolescent substance use, meaning that parental support can deter or
decrease substance use among adolescents (Barnes, Reifman, Farrell, & Dintcheff, 2000;
Chaplin et al., 2012; Wills & Cleary, 1996; Wills, Resko, Ainette, & Mendoza, 2004). Parental
support also has an indirect relationship with adolescent substance use via mediator variables.
One study found that the effects of parental support on adolescent alcohol and marijuana use
were mediated by self-control and risk-taking tendency (Wills, Resko, Ainette, & Mendoza,
2004). The researchers explained that parental support increased good self-control and decreased
risk-taking behaviors, which ultimately led to a decreased alcohol and marijuana use among
adolescents in the study. Another study found that parental support increased behavioral coping
skill, academic competence and decreased deviant-prone attitudes, which finally deterred or
decreased adolescent substance use (Wills & Cleary, 1996). Lack of parental support may
decrease close parent-child relationship, which consequently result in an initiation or an increase
in substance use (Chaplin et al., 2012).
Regarding components of parental support, Wills & Cleary (1996) concluded that
emotional support (e.g., adolescents share feelings with parents and parents listen to their
feelings) had stronger effects on adolescent substance use than instrumental support (e.g., parents
23
help with homework or help adolescents go somewhere). In tandem with its preventive effects,
parent support also has positive impact on substance abuse treatment outcomes. In a longitudinal
study with adolescents undergoing substance abuse treatment, Whitney, Kelly, Myers, and
Brown (2002) found that higher parental support was associated with lower levels of adolescent
drug and alcohol use during three and six-month follow-ups. In the relationship with parental
monitoring, parental support has indirect impact on adolescent substance use through parental
monitoring. Barnes, Reifman, Farrell, and Dintcheff (2000) concluded that children who are
reared in supportive and nurturing families were more likely to be receptive with parental
monitoring, which consequently drank less and had fewer times drunk.
2.3.3 Parental monitoring
Parental monitoring refers to the extent to which parents are aware of their children's activities
and whom they are with when not at home or in school, and the ultimate goal of parental
monitoring is to promote adolescents’ self-regulatory behaviors (DiClemente et al., 2001).
Numerous studies have confirmed that parental monitoring prevents adolescents from initiating
alcohol and marijuana use and is associated with their lower levels of substance use (Bahr,
Hawks, & Wang, 1993; Barnes, Reifman, Farrell, and Dintcheff, 2000; Van der Vorst, Engels,
Meeus, and Dekovic, 2006; DiClemente et al., 2001; Steinberg & Fletcher, 1994). Adolescents
who have high level of parental monitoring are less likely to initiate to use substances (Steinberg
& Fletcher, 1994; Barnes, Reifman, Farrell, and Dintcheff, 2000). Conversely, youth with little
or lack of parental monitoring are more likely to drink heavily and abuse drugs than those who
are closely monitored by their parents (Jessor, 1976; Fraser, 1984). In addition, parental
monitoring has an indirect impact on adolescent substance use through choices of peers. Teens
24
are much less likely to choose friends who use drugs when their parental monitoring is high
(Bahr, Hawks, & Wang, 1993).
2.3.4 Parental disapproval
Parental disapproval is measured by how parents would feel if their children drink or use drugs
as reported by their adolescent children (Johnston, O’Malley, & Bachman, 2001). Studies have
confirmed that parental disapproval is a protective factor for adolescent substance use, indicating
that higher levels of parental disapproval are associated with lower frequency of substance use
and greater likelihood of abstinence among adolescents (Sawyer & Stevenson, 2008; Donovan,
2004; Mrug & McCay, 2013; Martino, Ellickson, & McCaffrey, 2009). The effects of parental
disapproval on adolescent substance use are strongly correlated with peer disapproval. In a recent
study examining the effects of parental and peer disapproval on adolescent alcohol use, Mrug &
McCay (2013) found that although youth often received higher parental disapproval than peer
disapproval throughout adolescence, peer disapproval was stronger than parental disapproval;
and the combination of strong parental and peer disapproval was associated with the greatest
likelihood of abstinence and lowest level of alcohol use. According to this study, parental
disapproval is not enough, and thus, it needs to incorporate with peers to ensure the effectiveness
of substance use prevention among adolescents. In addition to direct impact, parental disapproval
also has indirect effects on adolescent substance use. Nash, McQueen, & Bray (2005) revealed,
in their longitudinal study, that students with more parental disapproval reported having greater
self-efficacy for avoiding alcohol use, fewer friends that drank alcohol, less approval for alcohol
use among close friends, and less alcohol use than those who reported some parental disapproval.
25
2.3.5 Race
In a cross-sectional study with White, Hispanic, and African American adolescents, Pilgrim,
Schulenberg, O'Malley, Bachman, and Johnston (2006) found that direct effects of parental
involvement on adolescent substance use was significant across all races, but the strength of the
effects was lower among African Americans than White and Hispanic counterparts. Similarly,
the strength of indirect effects, which were mediated by school success and time with friends,
was also lower among African Americans than White and Hispanic teens. With regard to
parental disapproval, racial differences are inconsistent among studies. Two studies found that
White adolescents received higher parental disapproval than Black counterparts (Mrug &
McCay, 2012; Catalano et al., 1992). Poverty, single parenthood, and community disadvantage
may be the main factors that result in lower perceptions of parental disapproval among Black
adolescents (Mrug & McCay, 2013). Contrary to the findings of the previous researchers, Foley,
Altman, Durant, & Wolfson (2004) did not find a significant difference in parental disapproval
among Black, White, and Hispanic adolescents. These inconsistent findings require further
investigation from researchers.
2.3.6 Age
Studies have found that parental involvement is more effective to deter or decrease substance use
among younger adolescents than older ones (Goncy & Van Dulmen, 2010; Pilgrim, Schulenberg,
O'Malley, Bachman, and Johnston, 2006). However, shared communication even makes older
adolescents drink more than younger ones (Goncy & Van Dulmen, 2010). Regarding parental
disapproval, its effects on adolescent substance use significantly vary among studies. Some
researchers found that parental disapproval had a stronger effect on alcohol use in earlier versus
later adolescence (Reifman, Barnes, Dintcheff, Farrell, & Uhteg, 1998). Others concluded that
26
parental disapproval was stronger for abstinence, but not for frequency of alcohol use among
older adolescents than younger ones (Mrug & McCay, 2013). Meanwhile, Sawyer & Stevenson
(2008) did not find any significant differences in the influence of parental disapproval on drug
use intentions between sixth and eighth graders in their study.
2.3.7 Gender
Cross-sectional and longitudinal studies have found that parental monitoring is associated with
lower level of substance use and is more effective in preventing adolescent substance use for
both boys and girls (Van der Vorst, Engels, Meeus, and Dekovic, 2006; Steinberg & Fletcher,
1994). However, the effects of parental monitoring is stronger for boys than girls, which means
that boys drink less than girls do when their parents monitor their drinking behavior (Van der
Vorst, Engels, Meeus, and Dekovic, 2006). Parental monitoring is less effective for male
substance users when peer influence is involved. Steinberg & Fletcher (1994) concluded that for
drug-using boys, their pattern of use was not influenced by levels of parental monitoring, but the
pattern of peer use. Concerning the effects of parental disapproval on adolescent substance use,
girls receive higher level of parental disapproval than boys throughout adolescence (Mrug &
McCay, 2013). But the protective effect of parental disapproval on early adolescents’ alcohol use
was stronger in boys than in girls (Kelly et al., 2011).
2.3.8 Summary and evaluation
Parental influence makes a significant contribution to deterring or decreasing levels of adolescent
substance use. The existing studies have confirmed the deterrent effects of parental involvement,
parental monitoring, parental support, and parental disapproval on substance use. However,
shared communication between parents and their children instigates adolescent substance use.
When examining the effects of these parental variables, researchers need to take into
27
consideration the impact of other variables such as school performance, self-control, and
especially peer influence. Future research should further examine racial differences in the effects
of parental variables on adolescent substance use as well as the impact of shared communication
between adolescents and their parents.
2.4 PEER INFLUENCE AND ADOLESCENT SUBSTANCE USE
A large body of research has revealed that peers have a strong influence on the development of
adolescent substance use; adolescents who have substance-using friends are more likely to use
substances (Branstetter, Low, & Furman, 2011; Epstein, Botvin, Baker, & Diaz, 1999; Maxwell,
2002; Marshal & Chassin, 2000; Wills & Cleary, 1999). Youth are more likely to increase their
frequency and levels of substance use commensuration with that of their peers (Ali & Dwyer,
2010; Branstetter, Low, & Furman, 2011). Friends do not only provide immediate access to
substances but also model substance-using behavior and shape positive attitudes toward the use
of substances (Farrell & White, 1998; Bray, Adams, Getz, & McQueen, 2003). Getting involved
with substance-using friends is a risk factor for the development of adolescent substance use.
Studies have shown that the more involved with substance-using friends the more likely youth
are to use substances or to increase their levels of substance use (Bahr & Hoffmann, 2008;
Moon, Blakey, Boyas, Horton, & Kim, 2014). In line with initiating and increasing levels of
substance use, the number of substance-using friends that youth have is also strongly related to
treatment outcomes. In their longitudinal study, Ramirez, Hinman, Sterling, Weisner, &
Campbell (2012) concluded that youth with less than four friends who use alcohol and drugs
were more likely to be abstinent than those with four or more friends who use the substances.
Whereas having peers who are less involved in substance use makes non-substance using
adolescents less likely to become a substance user (Steinberg & Fletcher, 1994). Adolescents are
28
more likely to be influenced by friends who are popular among their peers and those who are
significantly more popular than themselves (Tucker, de la Haye, Kennedy, Green, & Pollard,
2014). Nowadays, in addition to face-to-face interaction – the most influential way (Branstetter,
Low, & Furman, 2011), the prevalence of internet also makes a significant contribution to peer
influence on adolescent substance use. In a recent longitudinal study, Huang et al. (2014) found
that adolescents with a greater number of friends who posted partying and drinking pictures of
themselves online were significantly more likely to use alcohol. However, studies have found
that adolescents who have good self-control and high levels of discipline are more resilient to
peer influence as they are less likely to adopt the values of substance-using peers or model their
substance use behaviors (Marshal & Chassin, 2000; Wills & Cleary, 1999).
The quality of peer friendships is one of the determinants leading to substance use among
adolescents. Investigators have found that conflict, hostility, and negative interactions in
friendships are associated with greater substance use among youth (Branstetter, Low, & Furman,
2011; Windle, 1994; Branstetter, Low, & Furman, 2011). Other determinants which
prospectively predict initiation of adolescent substance use include peer approval and the use of
substance (Trucco, Colder, & Wieczorek, 2011). In contrast, peer disapproval is a protective
factor for adolescent substance use (Mason, Mennis, Linker, Bares, & Zaharakis, 2014). Studies
have concluded that peer disapproval is significantly associated with a decreased substance use
among adolescents (Mrug & McCay, 2013; Mason, Mennis, Linker, Bares, & Zaharakis, 2014;
Sawyer & Stevenson, 2008). When examining the effects of peer disapproval on adolescent
substance use, it is important to take parental disapproval into account because parental
disapproval increases peer disapproval and creates greater self-efficacy for avoiding substance
use (Nash, McQueen, & Bray, 2005). Since parental disapproval amplifies the protective effect
29
of peer disapproval, the combination of parental disapproval and peer disapproval is strongly
associated with an increased likelihood of abstinence and a decreased likelihood of frequent
substance use (Mrug & McCay, 2013). In tandem with parental disapproval, individual
religiosity is also a significant contributor to the impact of peer disapproval on adolescent
substance use. Bahr & Hoffmann (2008) revealed that a highly religious adolescent whose
friends used marijuana was less likely to use marijuana than an unreligious adolescent whose
friends used marijuana.
Studies on peer support reveal inconsistent findings. Some found that peer support was
associated with lower levels of substance use (Scholte, van Lieshout, & van Aken, 2001; Windle,
1994). Whereas others argued that peer support was associated with greater substance use among
adolescents (Averna & Hesselbrock, 2001; Wills, Resko, Ainette, & Mendoza, 2004). One
possible explanation for this counter-intuitive finding is that adolescents tend to select friends
with similar interests, values, beliefs, and attitudes (Youniss & Smoller, 1985), and thus, support
from substance-using friends may result in greater substance use among adolescents (Averna &
Hesselbrock, 2001).
The direction of the relationship between peer influence and adolescent substance use is
complex and varies significantly among studies. Two studies have found that the relationship
between peer influence and adolescent substance use is bidirectional, indicating that levels of
peer substance use are strongly associated with that of adolescent substance use and vice versa
(Curran, Stice, & Chassin, 1997; Bray, Adams, Getz, & McQueen, 2003). However, these two
longitudinal studies, which used similar analytical method - Latent Growth Analysis, revealed
two opposite directions of this relationship. In their study with White and Hispanic samples,
Curran, Stice, & Chassin (1997) found that higher levels of initial peer alcohol use was related to
30
larger increases in adolescent drinking. In contrast, Bray, Adams, Getz, & McQueen (2003), who
conducted the study with African American, Mexican American, and non-Hispanic White
adolescents, concluded that higher levels of initial peer drinking were related to smaller increases
in youth drinking. Whereas other researchers found that the relationship between peer influence
and adolescent substance use was unidirectional, meaning that adolescent alcohol use predicted
peer alcohol use rather than vice versa (Farrell, 1994; Farrell & Danish, 1993). However, this
finding from Farrell and Danish (1993) and Farrell (1994) exposes some limitations. First,
sample in their studies was exclusively African Americans whose peer influences have been
reported to be weaker than other ethnic groups (Mrug & McCay, 2013; Mason, Mennis, Linker,
Bares, & Zaharakis, 2014; Farrell & White, 1998). Second, the analyses of the studies consisted
of traditional fixed-effects autoregressive (AR) structural equation models, which do not take
into account growth or individual differences in growth over time (Rogosa, 1987).
2.4.1 Race
The extant studies on peer influence and adolescent substance use mainly focus on racial
differences between white and non-white (African American and Hispanic) populations.
Researchers have shared a common finding that peer influence is more strongly related to both
abstinence and frequency of substance use among White adolescents and less strongly related
among African American and Hispanic counterparts (Mrug & McCay, 2013; Mason, Mennis,
Linker, Bares, & Zaharakis, 2014; Farrell & White, 1998). However, this racial difference
disappears by late adolescence (Mrug & McCay, 2013). The strength of peer influence on Asian
American adolescents is still unknown in the existing literature, which needs more attention from
researchers.
31
2.4.2 Age
The effects of peer influence on adolescent substance use largely depend on their age. It is
commonly established that peer influence on substance use is predominant and stronger among
older adolescents than the younger ones (Mrug & McCay, 2013; Sawyer & Stevenson, 2008; Ali
& Dwyer, 2010; Branstetter, Low, & Furman, 2011).
2.4.3 Gender
Investigators have posited that having many substance-using friends makes it more likely for
boys to begin using substances or move from experimenters to heavy users than girls; and low
substance use by friends makes girls more likely to stop experimenting with substance use and
maintain their sobriety (Steinberg & Fletcher, 1994). Peers’ attitudes have stronger effects on
adolescent girls than adolescent boys, indicating that adolescent girls are less likely to use
substances if they receive unfavorable attitudes toward substance use from friends than boys
(Mason, Mennis, Linker, Bares, & Zaharakis, 2014). Adolescent girls receive more peer support
than boys (Wills, Resko, Ainette, & Mendoza, 2004) and peer disapproval is more influential for
them than boys (Mrug & McCay, 2013).
2.4.4 Summary and evaluation
Similar to parental influence, peer influence has strong effects on adolescent substance use and
its effects are even stronger than parental influence (Mrug & McCay, 2013). The number of
substance-using friends, friends’ favorable attitudes toward substance use, and negative
friendships are key determinants leading to adolescent substance use. When examining the
effects of peer influence on adolescent substance use, it is critical to take into account such
factors as individuals’ self-control, discipline, parental disapproval, peer disapproval, and
individual religiosity as they are correlated with peer influence and can protect adolescents from
32
using substances. Controversial findings about the effects of peer support suggest further
research on the nature of friendships and how adolescents select friends. It appears that there is a
reciprocal relationship between peer influence and adolescent substance use, and peer influence
is stronger for older adolescents. It is clear that peer influence is stronger among White
adolescents than non-white ethnic groups. Further research needs to examine racial differences
among other ethnic groups. Findings from the current studies address an important point that
prevention programs need to seek parental involvement, minimize their interactions with
substance-using friends, and maximize their peers’ unfavorable attitudes towards using
substances.
33
3.0 THEORETICAL FRAMEWORKS
3.1 SOCIAL LEARNING THEORY
3.1.1 Key concepts and assumptions
Social learning theory (Bandura, 1977) suggests continuous and reciprocal interaction between
the individuals’ cognition and behavior exist within the ecological environment where human
behavior is developed. According to Bandura (1977), human behavior is not inborn. Rather it is
learned through our socialization process. SLT utilizes key concepts such as observational
learning, imitation, modeling, and self-efficacy to explain the development of behavior.
Individual observational learning is acquired by attention to and retention of activities. Such
activities are determined by interpersonal interactions and behaviors of people with whom
individuals regularly associate. Imitation occurs when individuals want to convert their symbolic
behaviors into actions. Modeling is the stage where individuals have strong motivation to
deliberately shape their behaviors in accordance with symbolic behaviors of others. Self-efficacy
reflects the individuals’ ability to understand, evaluate, and alter their thinking, which allows for
differential responses to what is observed. According to the SLT, adolescents are vulnerable to
alcohol and drug use through regular observation and interaction with family and peers who use
substances. Regular observation and interaction enables adolescents attend to, memorize, and
want to imitate the substance use behavior.
3.1.2 Analysis of conceptual frameworks
Studies examining the relationship between parental substance use and their children’s substance
use reveal children whose parents frequently use drugs are more likely to use the substances than
children of parents who do not use drugs (Windle 2000; Drapela & Mosher, 2007; Miller,
Jennings, Alvarez-Rivera, & Miller, 2008). A similar relationship exists between sibling and peer
34
substance use. Adolescents who perceive benefits of alcohol and drug use from their elder
siblings are more likely to use the substances (Windle, 2000; Low, Shortt, & Snyder, 2012); and,
those who perceive greater peer approval of substance use are more likely to report lifetime
alcohol and marijuana use regardless of their own personal definitions (Miller, Jennings,
Alvarez-Rivera, & Miller, 2008). Previous research confirms both peer and family substance use
has direct effects on adolescent substance use (Windle, 2000; Bahr, Hoffmann, and Yang, 2005;
HeavyRunner-Rioux & Hollist, 2010).
One of the most effective applications of SLT is the use of peer educators as positive role
models for adolescents. According to Wodarski (2010), the Teams, Games, and Tournaments
treatment programs, combined with family therapy, anger management, and alcohol and drug
abuse education, have been effective in helping adolescents reduce their level of alcohol use.
This theory-driven treatment method gave participants an opportunity to learn positive behaviors
from their peers which subsequently reduced their substance use (Wodarski, 2010).
3.2 PROBLEM BEHAVIOR THEORY
3.2.1 Key concepts and assumptions
According to Jessor and Jessor (1977), the Problem Behavior theory is formulated by three
systems including (1) the personality system, (2) the perceived environment system, and (3) the
behavior system. Each of these systems is composed of variables that serve either as instigation
for engaging in problem behavior or controls against involvement in problem behavior (See
Figure 1).
3.2.1.1 The personality system
The personality system consists of three component structures – the motivational instigation
structure, the personal belief structure, and the personal control structure. The motivational
35
instigation structure is about the directional orientation of action, which is associated with both
value placed on goals and the expectation of attaining goals. Achievement of a goal largely
depends on value placed on the goal as value determines the direction of action to achieve the
goal. There are three central and salient goals for school-age adolescents including academic
achievement, independence, and peer affection. These goals comprise seven variables in the
motivational instigation structure – value on academic achievement, value on independence,
value on affection, expectation for academic achievement, expectation for independence,
expectation for affection, and the independence-achievement value discrepancy. The
independence-achievement value discrepancy refers to the degree to which the goal of
independence is valued more highly than the goal of academic achievement. The next component
of personality system is the personal belief structure, which refers to cognitive-control variables
exerted against the occurrence of problem behavior. These variables include social criticism,
alienation, self-esteem, and internal-external locus of control. Social criticism refers to the
degree of acceptance or rejection of the values, norms, and practices of the large society.
Alienation refers to a sense of uncertainty about self, a concern about one’s roles, and a belief
about isolation from involvement with others. High self-esteem can protect one from engaging in
problem behaviors. Internal locus of control reflects one’s commitment to the ideology of the
larger society. External locus control is a function to safeguard conventional behavior and protect
against nonconformity. Similar to the personal belief structure, the personal control structure also
refers to controls against non-normative behaviors. However, the difference between the
personal belief structure and the personal control structure is that variables in the personal belief
structure do not directly relate to behavior. Whereas variables in the personal control directly
link to or refer to behaviors. The personal control structure consists of three variables –
36
attitudinal tolerance of deviance, religiosity, and the discrepancy between positive and negative
functions of behaviors. High attitudinal intolerance of deviance is a direct control against
problem behaviors. Involvement with religious beliefs, ideology, and activities leads to moral
sanctioning and general concern with transgression. Control over engaging problem behaviors is
attenuated when positive functions outweigh negative functions. In the personality system,
problem behavior proneness includes lower value on academic achievement, higher value on
independence, greater social criticism, higher alienation, lower self-esteem, greater attitudinal
tolerance of deviance, and lower religiosity.
3.2.1.2 The perceived environment system
The perceived environment system consists of a distal structure and a proximal structure. The
distal structure is comprised of variables that do not directly or necessarily implicate problem
behaviors. In contrast, the proximal structure refers to variables that are directly or obviously
related to the occurrence of problem behaviors. The distal structure includes six variables –
perceived support from parents, perceived support from friends, perceived control from parents,
perceived control from friends, compatibility between parents and friends in their expectations
about a given adolescent, and the perceived influence on the adolescent from parents relative to
that from friends. High support and controls would protect adolescents from problem behaviors.
Compatibility refers to consensus between parents and friends’ expectations about the
adolescent. Low compatibility would result in a greater likelihood of the occurrence of problem
behaviors. The relative parent versus friends’ influence refers to the perception of greater past
and present influence from parents or friends. Parental influence is expected to be more
conventional than peer influence. Therefore, if adolescents receive less conventional standards
and have greater involvement of friends, both of these factors increase the likelihood of engaging
37
in problem behavior. The proximal structure refers to the prevalence of models and social
support for problem behavior. The prevalence of models implicates the opportunity to engage in
problem behaviors, and access to the problem behaviors (e.g., drug supply). Social support for
problem behaviors implies positive approval for involving in the behavior, social pressure from
others, and lack of disapproval from others. The proximal structure includes three main variables
– friends’ approval-disapproval of problem behavior, parental approval-disapproval of problem
behavior, and friends’ models for problem behavior. In summary, the perceived environment
system deals with both the perception of social controls against problem behavior and the
perception of models and support for problem behavior. Social controls are largely located with
the distal structure. Whereas models and support for problem behavior are located in proximal
structure. Theoretically, problem behavior proneness in the distal structure includes low parental
support and controls, low peer controls, low compatibility between parent and peer expectations,
and low parental influence and high peers influence. In the proximal structure, problem behavior
proneness is characterized by low parental disapproval of problem behavior, and high peers
models and approval for engaging in problem behavior.
3.2.1.3 The behavior system
The behavior system implicates to the structure of problem behavior and the structure of
conventional behavior. The problem structure refers to adolescents’ inappropriate or undesirable
actions as considered by the larger society. Whereas the conventional behavior structure refers to
socially approved and normatively expected behaviors. The problem behavior structure includes
marijuana use, sexual intercourse, activism or social protest behavior, drinking, drinking
problem, general deviant behavior, and multiple problem behavior. The conventional behavior
structure is comprised of two variables – religious involvement as measured by frequency of
38
church attendance and religious activities, and involvement with academic course work and
achievement as measured by grade-point average. Problem behavior proneness in the behavior
system includes high involvement in other problem behaviors and low involvement in
conventional behaviors.
3.2.2 Analysis of conceptual frameworks
Jessor and Jessor (1977) tested their theoretical frameworks with over 400 high school students
and 200 college students in a four-year longitudinal study from 1969 to 1972. With regard to the
personality system, Jessor and Jessor (1977) found that personal controls had the most direct and
substantial relationship with problem behavior; motivational instigation was the next important
structure; personal beliefs, however, were least connected with problem behavior as only social
criticism variable was statistically significant with problem behavior. Results of the study also
revealed that in the perceived environment system, the proximal structure had the strongest
influence on problem behavior, especially peers approval variable and peers models of problem
behavior. Additionally, parental approval and lack of parental disapproval were also significantly
associated with problem behavior. For distal structure, both parental support and parental
controls deterred problem behaviors; specifically, the strength of parent support was stronger
than parent controls. Similarly, peers controls also had deterrent effects on problem behavior,
indicating that adolescents are less likely to get involved in problem behavior when they perceive
sanctions and criticism from friends. However, friends support was irrelevant to problem
behavior as the relationship was not statistically significant. Moreover, adolescents were at
greater risk of engaging in problem behavior when there was greater incompatibility between
parents and friends’ expectations and greater influence of friends.
39
Jessor and Jessor (1977) also concluded that adolescents who have higher attitudinal
intolerance about transgression and those who are more religious were less likely to engage in
problem behavior; those who have positive perceptions about drinking and using marijuana were
more likely to use the substances than those who have negative perceptions; adolescents who
place more value on academic achievement and have higher expectations on academic
achievement were less likely to get involved in problem behavior; those who have more value on
independence and more expectation to attain independence goal were at higher risk of problem
behavior proneness.
Conceptual frameworks of the Problem Behavior theory (Jessor and Jessor, 1977) have
been tested with adolescent alcohol and marijuana use and results have shown significant
correlations with the problems. Cross-sectional (Jessor, Chase, & Donovan, 1980) and
longitudinal (Jessor, 1987) studies have concluded that adolescent alcohol and marijuana use are
associated with lower value on academic achievement, higher value on independence, greater
attitudinal tolerance of deviance, lesser religiosity, less compatibility between parents and
friends, greater perceived influence from friends than parents, greater friends approval for
problem behavior, greater friends models for problem behavior, greater involvement in other
problem behavior, and less involvement with conventional behavior such as attending church.
The researchers have also found that adolescent drinking problem is predictive of adolescent
marijuana use and vice versa (Jessor, Chase, & Donovan, 1980; Jessor, 1987). So far, the
Problem Behavior theory has been tested in numerous studies in both the United States (De Leo
& Wulfert, 2013; Mobley & Chun, 2013) and oversea countries (Ndugwa et al, 2011; Jessor,
Turbi, Costa, Dong, Zhang, and Wang, 2003). Results of these studies have confirmed the
conceptual frameworks of the theory that protective factors (support, control, and models) and
40
risk factors (models, vulnerability, and opportunity) significantly prevent/ decrease or predict the
development of problem behaviors (alcohol problems, marijuana use, cigarette smoking, and
risky sexual intercourses) among adolescents (Ndugwa et al, 2011; Jessor, Turbi, Costa, Dong,
Zhang, and Wang, 2003); those who have greater attitudinal tolerance of deviance and have less
value on academic achievement are more likely to use drugs, smoke cigarette, and have risky
sexual behaviors (De Leo & Wulfert, 2013).
Figure 1. The conceptual structure of Problem Behavior Theory
Figure 1. The conceptual structure of Problem Behavior Theory
41
4.0 METHODOLOGY
4.1 STUDY DESIGN AND PROCEDURE
The 2013 National Survey on Drug Use and Health (NSDUH) is the 33rd in a series of survey
conducted by the Federal Government since 1971 and is sponsored by SAMHSA, U.S.
Department of Health and Human Services. The primary purpose of this survey is to measure the
prevalence and correlates of drug use in the United States. This survey series provide
information about the use of tobacco, alcohol, marijuana, cocaine, crack cocaine, heroin,
hallucinogens, inhalants, pain relievers, tranquilizers, stimulants, and sedatives among the non-
institutionalized United States civilian population aged 12 and older in 50 States and the District
of Columbia. This is the best way to estimate different types of drug use virtually in the entire
the United States. The 2013 NSDUH is a cross-sectional study because participants’ interview
were only conducted one time. Therefore, the survey only provides an overview of the
prevalence of drug use at in 2013 rather than a view of how drug use changes over time for
specific individuals.
The 2013 NSDUH used computer-assisted interviewing (CAI) methods, which combined
computer-assisted personal interviewing (CAPI) conducted by an interviewer and audio
computer-assisted self-interviewing (ACASI) for data collection. Usage of ACASI is to provide
respondents with a highly private and confidential means of responding to questions and to
increase the level of honest reporting of illicit drug use and other sensitive behaviors. To collect
information, field interviewers visited each sample address to determine dwelling unit eligibility,
to select participants, and to conduct interviews. The interviewers also identified and
immediately followed any new housing units or any dwelling units missed during the advance
listing process. The interviewers used a portable computer to do screening process, select
42
participants, and conduct interviews with eligible participants at their homes. A total of 67,838
CAI interviews was obtained in 2013, and 83.93% of them responded to the questionnaires. The
data was weighted to obtain unbiased estimates for survey outcomes. Throughout the course of
the study, participants’ anonymity and privacy of responses were protected by hiding identifying
information from survey responses in compliance with Federal laws. In addition, questionnaires
of the survey and the interviewing procedures were designed to enhance the privacy of
responses. ACASI was used to gather answers to sensitive questions. Each participant who
completed a full interview was given a $30 cash payment as a token of appreciation for his or her
time.
4.2 PARTICIPANTS FOR THE CURRENT STUDY
The scope of this study aims at White, African American, and Asian American adolescents aged
12 to 17 years old. A total of 12,984 adolescents was computed from the 2013 NSDUH data. The
majority of the total samples are White adolescents, N= 9,920 (76.4%), which is followed by
African American youth, N= 2,420 (18.6%). Meanwhile, Asian American counterparts take the
smallest proportion of the samples, N= 644 (5.0%). In this study, male adolescents are 51.0%
(N= 6,618). Participants aged 12, 13, 14, 15, 16, and 17 accounted for 1,952 (15%), 2,111
(16.3%), 2,215 (17.1), 2,232 (17.2%), 2,248 (17.3%), and 2,226 (17.1%) respectively.
4.3 MEASURES
4.3.1 Dependent variables
4.3.1.1 Marijuana use.
The participants were asked if they had ever used marijuana. Their answers were coded as 0=
“No” and 1= “Yes or used”.
43
4.3.1.2 Alcohol use.
The participants were asked if they had ever used alcohol. Their answers were coded as 0= “No”
and 1= “Yes or used”.
4.3.2 Predictors
4.3.2.1 Religiosity.
This latent construct examines how religious the participants are. This score variable was created
by summing Z scores of 5 individual religiosity items: (1) Number of religious services you
attended in the past 12 months (2) Number of church or faith-based activities you attended in the
past 12 months, (3) My religious beliefs are very important (4), My religious beliefs influence
my decisions, and (5) It is important that my friends share religious beliefs. The Cronbach alpha
of these items yielded at 0.83, and the higher scores indicate higher levels of religiosity. The
distributions of this variable are as follows: M (.0019), SD (.77278), Skewness (.006), and
Kurtosis (-.841).
4.3.2.2 School-based prevention programs.
This variable examines if the participants have attended a special class about drugs and alcohol
in school during the past 12 months. The participants’ answers were recoded as 0= “No” and 1=
“Yes”.
4.3.2.3 Parental support.
This variable examines the participants’ perceptions about emotional support and help in study
that they received from their parents. It combines four specific items: (1) How often did your
parents check if you have done your homework? (2) How often did your parents provide help
with your homework when you needed it? (3) How often did your parents let you know you have
done a good job? and (4) How often did your parents tell you they were proud of you for
44
something you had done? The standardized Cronbach’s alpha of these items was .77. In the
original data set, the answers of these items were coded as 1= “Always”, 2= “Seldom”, 3=
“Sometimes”, and 4= “Never”. In this study, these items were reversely recoded as 0= “Never”,
1= “Seldom”, 2= “Sometimes”, and 3= “Always” so that they have the same direction with other
variables. The distribution of the Parental support variable ranges from 0 to 12, indicating that
the higher scores the more parental support the participants received. Due to high skewness of
the distribution, Parental support was dichotomized as “low parental support” for scores from 0
to 9) versus “high parental support” for scores from 10 to 12.
4.3.2.4 Parental monitoring.
Parental monitoring measures the participants’ perceptions about their parents’ monitoring on
their activities in the past 12 months. It is a combination of three continuous variables: (1) How
often parents limited the amount of time adolescents watched TV, and (2) How often parents
limited the amount of time adolescents went out with friends on school nights; and (3) How often
your parents made you do chores around the house. The standardized Cronbach’s alpha of these
items was .50. Similar to the measurements of Parental support variable, these two variables
were recoded as 0= “Never”, 1= “Seldom”, 2= “Sometimes”, and 3= “Always”. The distribution
of Parental monitoring ranges from 0 to 9, indicating that the higher scores the higher parental
monitoring the participants receive. Due to its high skewness, Parental monitoring was
dichotomized as “low parental monitoring” for scores from 0 to 5 versus “high parental
monitoring” for scores from 4 to 6.
4.3.2.5 Parental disapproval.
Parental disapproval measures how adolescents feel about their parents’ attitudes towards their
drinking and marijuana use. It consists of three specific items: (1) How do you think your parents
45
would feel about you trying marijuana or hashish once or twice? (2) How do you think your
parents would feel about you using marijuana or hashish once a month or more? and (3) How do
you think your parents would feel about you having one or two drinks of an alcoholic beverage
nearly every day? The scales of these items include: 1= “Neither disapprove nor approve”, 2=
“Somewhat disapprove”, and 3= “Strongly disapprove”. The standardized Cronbach’s alpha of
these items yielded at .84. The distribution of Parental disapproval ranges from 3 to 9, meaning
that the higher scores the more parental disapproval the participants received. Due to the issue of
normality distribution, Parental disapproval was dichotomized as 2= “Strong parental
disapproval” for score of 9, and 1= “Everyone else” for scores from 3 to 8.
4.3.2.6 Peer substance use.
Peer substance use examines the proportion of alcohol and marijuana use among the participants’
class mates. It was created by summing up three specific items: (1) How many of the students in
your grade at school would you say use marijuana or hashish? (2) How many of the students in
your grade at school would you say drink alcoholic beverages? and (3) How many of the
students in your grade at school would you say get drunk at least once a week? The
measurements of these items include: 1= “None of them”, 2= “A few of them”, 3= “Most of
them”, and 4= “All of them”. The standardized Cronbach’s alpha of these items was .87. The
distribution of Peer substance use ranges from 3 to 12, indicating that the higher scores the more
substance- using friends that the participants had. The distributions of this variable are as
follows: M (5.85), SD (2.07), Skewness (.176), and Kurtosis (-.809).
4.3.2.7 Peer disapproval.
Peer disapproval examines the participants’ perceptions about their close friends’ attitudes
towards their drinking and marijuana use. It was created by adding up three items: (1) How do
46
you think your close friends would feel about you trying marijuana or hashish once or twice? (2)
How do you think your close friends would feel about you using marijuana or hashish once a
month or more? and (3) How do you think your close friends would feel about you having one or
two drinks of an alcoholic beverage nearly every day? The scales of these items have three
levels: 1= “Neither disapprove nor approve”, 2= “Disapprove”, and 3= “Strongly disapprove”.
Their standardized Cronbach’s alpha yielded at .89. The distribution of Peer disapproval ranges
from 3 to 9, indicating that the higher score, the more peer disapproval participants received. Due
to its high skewness, Peer disapproval was dichotomized as 2= “Strong peer disapproval” for
score of 9, and 1= “Everyone else” for scores of 3 to 8.
4.3.2.8 Race.
In the 2011 NSDUH data set, this categorical variable consisted of seven categorizations. In this
study, only three of them were used including White, African American, and Asian adolescents,
which were recoded as 1= “Whites”, 2= “African Americans”, and 3= “Asian Americans”.
4.3.2.9 Age.
Originally, this continuous variable included all participants aged 12 to 65 or older. However,
only adolescents aged 12 to 17 were selected for this study.
4.3.2.10 Gender.
This categorical variable was recoded as 1= “Male” and 2= “Female”.
4.4 DATA ANALYSES
Preliminary analyses were conducted to check normality distributions, and bivariate relationships
of all predictors and dependent variables. The purpose of preliminary analyses aimed to check
47
assumptions of this study. Specifically, preliminary analyses checked skewness, kurtosis, means,
medians, modes, standard deviations, and Chi square or F tests.
Since predictors are either categorical or continuous variables, while alcohol and
marijuana use are dichotomous, binary logistic regression analyses were computed to examine
the odds of marijuana and alcohol use occurring as the values of religiosity, school-based
prevention programs, parental support, parental monitoring, parental disapproval, peer use, and
peer disapproval variables change, controlling for demographic variables (age, race, and gender).
Then, these analyses were followed by moderation and mediation analyses to explore (1) if the
relationship between religiosity and alcohol and marijuana use among these adolescents is
moderated by age, race, and gender; and (2) if religiosity acts as a mediator of the presumed
tendency for younger, Black, and female youth to use marijuana and alcohol.
48
5.0 RESULTS
5.1 DESCRIPTIVE ANALYSES OF THE STUDY VARIABLES
Descriptive statistics (see Tables 1 and 2) show that the percentages of those who did not use
marijuana and alcohol were much higher than that of those who used the substances; the majority
of participants in the study were whites followed by African Americans and then Asian
Americans; the percentages of male and female participants were almost equal; similarly, the
percentage of those who received low parental monitoring was almost the same as the percentage
of those who received high parental monitoring; those who received high parental support
outnumbered those who received low parental support; the majority of participants received
parental disapproval for their substance use; the majority of participants received peer
disapproval for their substance use; the percentage of those who attended a special class on drugs
and alcohol use were less than those who did not attend the class; the participants’ number of
substance-using friends ranged from three to 12; and the participants’ religiosity scored from -
1.61 to 1.54. (For more details regarding distributions of variables in the study see Table 1 and
2).
49
Table 1. Descriptive Analysis of Marijuana Use, Alcohol Use, Race, Gender, Parental Monitoring, Parental Support,
Parental Disapproval, Peer Disapproval, and School-Based Prevention Programs Variables (N= 12,984)
Variables n %
Marijuana use
No
Yes
Alcohol use
No
Yes
Race
White
African American
Asian American
Gender
Male
Female
Parental monitoring
Low parental monitoring
High parental monitoring
Parental support
Low parental support
High parental support
Parental disapproval
Everyone else
Parental disapproval
Peer disapproval
Everyone else
Peer disapproval
Special class on drugs and alcohol use
No
Yes
10790
2184
8887
4088
9920
2420
644
6618
6366
5959
6062
4937
7196
2276
10522
5568
7181
6939
5204
83.1
16.8
68.4
31.5
76.4
18.6
5.0
51.0
49.0
45.9
46.7
38.0
55.4
17.5
81.1
42.9
55.3
53.4
40.1
Table 2. Descriptive Analysis of Age, Number of Substance-Using Friends, and Religiosity Variables (N=12,984)
Variables M (SD) Min/Max Potential Scores
Age
Number of substance-using friends
Religiosity
14.5692 (1.68747)
5.8506 (2.06855)
.0019 (.77278)
12.00 – 17.00
3.00 – 12.00
-1.61 – 1.54
12.00 – 17.00
3.00 – 12.00
-1.61 – 1.54
50
5.2 BIVARIATE ANALYSES
5.2.1 Bivariate analysis of all variables
Results of bivariate correlations analysis (Table 3) revealed that individual religiosity was
statistically significant with both alcohol and marijuana use, indicating higher religiosity is
associated with less alcohol and marijuana use among youth. With regards to the relationships
between demographic variables and religiosity, the findings show that African American youth
are more religious than white counterparts; there is no significant difference in religious beliefs
between white and Asian American adolescents; younger youth are more religious than older
ones; and female gender are more religious than male gender.
Table 3. Bivariate Analysis of all Predictors and Outcome Variables (N=12,984)
Variables 1 2 3 4 5 6 7 8 9 10 11 12 13
1. AA -- -.109** .028** .000 .085** -.002 .038** .003 .071** -.035** -.035** -.013 .044**
2. Asian -.109** -- .002 .007 -.002 -.047** .005 .040** -.041** .038** .027** -.064** -.046**
3. Age .028** .002 -- .008 -.123** -.223** -.127** -.209** .605** -.350** -.157** .413** .336** 4. Gender .000 .007 .008 -- .085** -.046** .028** .046** .107** .078** .027** .007 -.026**
5. Religiosity .085** -.002 -.123** .085** -- .173** .176** .240** -.154** .242** .025** -.195** -.213**
6. Parent Sup -.002 -.047** -.223** -.046** .173** -- .209** .147** -.215** .232** .071** -.214** -.178**
7. Parent Mo .038** .005 -.127** .028** .176** .209** -- .141** -.127** .150** .068** -.139** -.110**
8. Parent Dis .003 .040** -.209** .046** .240** .147** .141** -- -.233** .407** .056** -.317** -.361**
9. Peer Use .071** -.041** .605** .107** -.154** -.215** -.127** -.233** -- -.414** -.079** .449** .387** 10. Peer Dis -.035** .038** -.350** .078** .242** .232** .150** .407** -.414** -- .065** -.413** -.400**
11. Program -.035** .027** -.157** .027** .025** .071** .068** .056** -.079** .065** -- -.105** -.091**
12. Al Use -.013 -.064** .413** .007 -.195** -.214** -.139** -.317** .449** -.413** -.105** -- .531** 13. Mari Use .044** -.046** .336** -.026** -.213** -.178** -.110** -.361** .387** -.400** -.091** .531** --
*p<.05; **p<.01
5.2.2 Bivariate analysis predicting marijuana use by all predictors
Table 4 represents a bivariate analysis of the variables race, gender, parental monitoring, parental
support, parental disapproval, peer attitudes, and school-based prevention programs by the
dependent variable – marijuana use. Chi-square analysis indicated that race statistically predicted
marijuana use among white, African American, and Asian American adolescents, χ2(2)=48.361,
p<.001. Specially, the percentages of white, African American, and Asian American adolescents
who used marijuana were 16.5% (N=1634), 20.3% (n=491), and 9.2% (N=59) respectively. Chi-
square analyses also indicated that more male adolescents (17.8%; n=1177) used marijuana than
51
female counterparts (15.8%; N=1007), χ2(1)=9.059, p<.01; a significantly higher percentage of
those who received low parental monitoring used marijuana (21.7%; N=1290) in comparison
with the percentage of those who received high parental monitoring (13.3%; N=808),
χ2(1)=144.582, p<.001; those who received low parental support used marijuana (25.3%;
N=1249) two times more than those who received high parental support (11.6%; N=837),
χ2(1)=384.637, p<.001; those who received strong parental disapproval used marijuana (10.5%;
N=1109) much less than everyone else (45.8%; N=1043), χ2(1)=1665.296, p<.001; similarly, a
significantly lower percentage of those who received strong peer disapproval used marijuana
(3.7%; N=265) relative to the percentage of everyone else (33.9%; N=1885), χ2(1)=2035.493,
p<.001; and those who did not attended a special class on drugs and alcohol used marijuana
(20.3%; N=1406) more than those who attended the drug and alcohol class (13.3%; N=691),
χ2(1)=101.539, p<.001.
52
Table 4. Bivariate Analysis Predicting Marijuana Use by Race, Gender, Parental Monitoring, Parental Support,
Parental Disapproval, Peer Disapproval, and School-Based Prevention Programs Variables (N=12,984)
Have You Used Marijuana?
Variables Yes No
n % n % X2(df) Cramer’sV
Race
White
African American
Asian American
1634
491
59
(16.5)
(20.3)
(9.2)
8281
19275
82
(83.5)
(79.7)
(90.8)
48.361(2)*** .061***
Gender
Male
Female
11771
007
(17.8)
(15.8)
54345
356
(82.2)
(84.2)
9.059(1)** .026**
Parental monitoring
Low parental monitoring
High parental monitoring
12908
08
(21.7)
(13.3)
46645
251
(78.3)
(86.7)
144.582(1)*** .110***
Parental support
Low parental support
High parental support
12498
37
(25.3)
(11.6)
36846
356
(74.7)
(88.4)
384.637(1)*** .178***
Parental disapproval
Everyone else
Parental disapproval
10431
109
(45.8)
(10.5)
12329
406
(54.2)
(89.5)
1665.296(1)*** .361***
Peer disapproval
Everyone else
Peer disapproval
Special class on drugs and alcohol use
No
Yes
18852
65
14066
91
(33.9)
(3.7)
(20.3)
(13.3)
36806
912
55294
510
(66.1)
(96.3)
(79.7)
(86.7)
2035.493(1)***
101.539(1)***
.400***
.091***
**p<.01, ***p<.001
Table 5 represents a bivariate analysis of the variables substance-using friends, age, and
religiosity by the dependent variable - marijuana use. Independent-sample t-tests indicated that
the mean score reflecting number of substance-using friends was higher among adolescents who
used marijuana (M=7.57, SD=1.64) relative to the mean score of those who did not use
marijuana (M=5.48, SD=1.96), t(3426.6)=-50.289, p<.001; the mean score of age was higher
among those who used marijuana (M=15.83, SD=1.20) than the mean score of those who did not
use the substance (M=14.31, SD=1.66), t(4082.8)=-50.282, p<.001; and the mean sore of
religiosity was much lower among those who used marijuana (M=-0.36, SD=0.70) than the mean
score of those who did not use the substance (M=0.08, SD=0.77), t(3289.1)=26.188, p<.001.
53
Table 5. Bivariate Analysis Predicting Marijuana Use by Substance-Using Friends, Age, and Religiosity
(N=12,984)
Variables
M (SD) t (df) p Pt. biserial
Have you used marijuana? (Peer Use)
Yes
No
Have you used marijuana? (Age)
Yes
No
Have you used marijuana? (Religiosity)
Yes
No
7.57 (1.64)
5.48 (1.96)
15.83 (1.20)
14.31 (1.66)
-.36 (.70)
.08 (.77)
-50.289 (3426.6)
-50.282 (4082.8)
26.188 (3289.1)
<.001
<.001
<.001
.387**
.336**
-.213**
5.2.3 Bivariate analysis predicting alcohol use by all predictors
Table 6 represents a bivariate analysis of the variables race, gender, parental monitoring, parental
support, parental disapproval, peer disapproval, and school-based prevention programs by the
dependent variable – alcohol use. Chi-square analysis indicated that race statistically predicted
alcohol use among white, African American, and Asian American adolescents, χ2(2)=58.224,
p<.001. Specially, the percentages of white, African American, and Asian American adolescents
who used alcohol were 32.7% (N=3238), 30.2% (n=731), and 18.5% (N=119) respectively. Chi-
square analyses also indicated that gender did not statistically predict alcohol use among the
study participants, χ2(1)=0.726, p>.05; those who received low parental monitoring used alcohol
(39.1%; N=2327) almost three times more than those who received high parental monitoring
(13.3%; N=808), χ2(1)=232.282, p<.001; a significantly higher percentage of those who received
low parental support used alcohol (44.2%; N=2128) relative to the percentage of those who
received high parental support (23.8%; N=2715), χ2(1)=557.029, p<.001; those received strong
parental disapproval used alcohol (24.7%; N=2601) much less than everyone else (63.2%;
N=1438), χ2(1)=1281.718, p<.001; similarly, those who received strong peer disapproval used
54
alcohol (14.7%; N=1054) almost four time less than everyone else (53.4%; N=2975),
χ2(1)=2178.682, p<.001; and those who attended a special class on drugs and alcohol used
alcohol (32.2%; N=3912) less than those who did not attend the drug and alcohol class (36.5%;
N=2531), χ2(1)=134.555, p<.001.
Table 6. Bivariate Analysis Predicting Alcohol Use by Race, Gender, Parental Monitoring, Parental Support,
Parental Disapproval, Peer Disapproval, and School-Based Prevention Programs Variables (N=12,984)
Have You Used Alcohol?
Variables Yes No
n % n % X2(df) Cramer’s V
Race
White
African American
Asian American
32387
31
119
(32.7)
(30.2)
(18.5)
66771
68652
4
(67.3)
(69.8)
(81.5)
58.224(2)*** .067***
Gender
Male
Female
20612
027
(31.2)
(31.9)
45524
335
(68.8)
(68.1)
.726(1) .007
Parental monitoring
Low parental monitoring
High parental monitoring
23278
08
(39.1)
(13.3)
36304
482
(60.9)
(74.0)
232.282(1)*** .139***
Parental support
Low parental support
High parental support
21821
715
(44.2)
(23.8)
27535
478
(55.8)
(76.2)
557.029(1)*** .214***
Parental disapproval
Everyone else
Parental disapproval
14382
601
(63.2)
(24.7)
837
7916
(36.8)
(75.3)
1281.718(1)*** .317***
Peer disapproval
Everyone else
Peer disapproval
Special class on drugs and alcohol use
No
Yes
29751
054
25313
912
(53.4)
(14.7)
(36.5)
(32.2)
25916
123
44058
226
(46.6)
(85.3)
(63.5)
(67.8)
2178.682(1)***
134.555(1)***
.413***
.105***
**p<.01, ***p<.001
Table 7 represents a bivariate analysis of the variables substance-using friends, age, and
religiosity by the dependent variable – alcohol use. Independent-sample t-tests indicated that the
mean score reflecting number of substance-using friends was higher among adolescents who
used alcohol (M=7.17, SD=1.74) relative to the mean score of those who did not use alcohol
(M=5.19, SD=1.90), t(8228.1)=-55.461, p<.001; the mean score of age was higher among those
who used alcohol (M=15.60, SD=1.20) than the mean score of those who did not use the
55
substance (M=14.10, SD=1.61), t(9303.6)=-54.937, p<.001; and the mean sore of religiosity was
much lower among those who used alcohol (M=-0.22, SD=0.74) than the mean score of those
who did not use the substance (M=0.10, SD=0.77), t(8124.5)=22.770, p<.001.
Table 7. Bivariate Analysis Predicting Alcohol Use by Substance-Using Friends, Age, and Religiosity (N=12,984)
Variables
M (SD) t (df) p Pt. biserial
Have you used alcohol? (Peer Use)
Yes
No
Have you used alcohol? (Age)
Yes
No
Have you used alcohol? (Religiosity)
Yes
No
7.17 (1.74)
5.19 (1.90)
15.60 (1.20)
14.10 (1.61)
-.22 (.74)
.10 (.77)
-55.461 (8228.1)
-54.937 (9303.6)
22.770 (8124.5)
<.001
<.001
<.001
.449**
.413**
-.195**
Since binary logistic regression does not assume linearity, normal distribution of scores, or
homoscedasticity, there is not a need to test for these assumptions. I have tested multicollinearity
assumption, which aims to check if there is a strong correlation among predictors in regression
models. The test result revealed that this assumption was met because its VIP value was less than
10.
5.3 BINARY LOGISTIC REGRESSION ANALYSIS PREDICTING MARIJUANA
USE BY RELIGIOSITY, SCHOOL-BASED PREVENTION PROGRAMS, PARENTAL
INFLUENCE, PEER INFLUENCE, AND DEMOGRAPHIC VARIABLES
To investigate how well religiosity and school-based prevention program influence marijuana
use among white, African American and Asian American adolescents, a binary logistic
regression analysis was conducted, employing marijuana use as an outcome variable.
Demographic variables including age, race, and gender were entered into the first block. The
second block included parental monitoring, parental support and parental disapproval. Parental
56
variables were followed by peer substance use and peer disapproval in the third block. The
independent variable – religiosity was entered in the fourth block, controlling for demographic,
parental, and peer variables. Finally, school-based prevention programs variable was added to
the fifth block to see if school-based prevention programs had deterrent effects on marijuana use
among the study participants, controlling for demographic variables, parental influence, peer
influence, and individual religiosity. The logic of entering variables into separate blocks is to test
the whole model and the separate sets of variables.
Table 8 represents a binary logistic regression analysis examining the relationships
between the predictors and marijuana use (No/Yes). Data indicated that the overall model was
statistically significant, χ2(11, N=12,984) = 3460.909, p<.001. Furthermore, data indicated that
85.4% of cases were categorized correctly. In terms of individual predictors, results of
demographic variables, χ2(4, N=12,984)= 1461.618, p<.001, showed that older adolescents were
almost 1.4 times (OR=1.366, 95% CI=1.300-1.435) more likely to use marijuana than younger
ones; African American adolescents were 1.45 times (OR=1.451, 95% CI=1.247-1.689) more
likely to use marijuana than white youths; Asian American adolescents were 1.4 times
(OR=.710, 95% CI=.511-.987) less likely to use marijuana than white counterparts; female
participants were 1.2 times (OR=.859, 95% CI=.761-.970) less likely to use marijuana than
males. Findings of parental set of variables, χ2(3, N=12,984) = 949.630, p<.001, indicated that
these who received high parental support in study was 1.3 times (OR=.758, 95% CI=.671-.857)
less likely to use marijuana than those who received low parental support; similarly, those who
received strong parental disapproval were almost 2.9 times (OR=.350, 95% CI=.308-.399) less
likely to use marijuana than everybody else; however, there was no significant difference
between “low parental monitoring” and “high parental monitoring” participants on marijuana
57
use. The third block, χ2(2, N=12,984) = 950.999, p<.001, reported that the more substance-using
friends the more likely the participants used marijuana (OR=1.404, 95% CI=1.350-1.460); those
who received strong peer disapproval were 4.6 times (OR= .219, 95% CI=.187-.256) less likely
to use marijuana than everyone else. Result of the independent variable - religiosity, χ2(1) =
86.318, p<.001, revealed that the more religious the less likely the adolescents used marijuana
(OR=.671, 95% CI=.617-.730). Result of the school-based prevention program model indicated
that those who attended a special class on drugs or alcohol were 1.2 times (OR=.814, 95%
CI=.718-.921) less likely to use marijuana than those who did not attend the prevention training
programs, χ2(1, N=12,984) = 10.595, p<.01.
Table 8. Binary Logistic Regression Analysis Examining Marijuana Use (N=12,984)
Variables
B (SE) Wald (X2) OR (95% CI)
Block 1
Age
African American
Asian American
Gender (1)
.312 (.025)
.372 (.077)
-.342 (.168)
-.152 (.062)
154.925***
23.157***
4.150**
6.022**
1.366 (1.300-1.435)
1.451 (1.247-1.689)
.710 (.511-.987)
.859 (.761-.970)
Block 2
Parental monitoring (1)
Parental support (1)
Parental disapproval (1)
.058 (.063)
-.277 (.062)
-1.049 (.066)
.829
19.615***
250.027***
1.059 (.936-1.200)
.758 (.671-.857)
.350 (.308-.399)
Block 3
Peer substance use
Peer disapproval (1)
.339 (.020)
-1.518 (.080)
291.985***
359.153***
1.404 (1.350-1.460)
.219 (.187-.256)
Block 4
Religiosity
-.399 (.043)
86.318***
.671 (.617-.730)
Block 5
Prevention programs
-.206 (.064)
10.543**
.814 (.718-.921)
Note: White is the reference group for race
For Model: R2 = .269 (Cox & Snell), R2 = .440 (Nagelkerke), χ2(11) = 3460.909, p<.001
Block 1: R2 = .124 (Cox & Snell), R2 = .203 (Nagelkerke), χ2(4)= 1461.618, p<.001
Block 2: R2 = .196 (Cox & Snell), R2 = .321 (Nagelkerke), χ2(3) = 949.630, p<.001
Block 3: R2 = .262 (Cox & Snell), R2 = .429 (Nagelkerke), χ2(2) = 950.999, p<.001
Block 4: R2 = .268 (Cox & Snell), R2 = .439 (Nagelkerke), χ2(1) = 88.067, p<.001
Block 5: R2 = .269 (Cox & Snell), R2 = .440 (Nagelkerke), χ2(1) = 10.595, p<.01
*p<.05, **p<.01, ***p<.001.
58
5.4 BINARY LOGISTIC REGRESSION ANALYSIS PREDICTING ALCOHOL USE
BY RELIGIOSITY, SCHOOL-BASED PREVENTION PROGRAMS, PARENTAL
INFLUENCE, PEER INFLUENCE, AND DEMOGRAPHIC VARIABLES
Table 9 represents a binary logistic regression analysis examining the relationships between the
predictors and alcohol use (No/Yes). Data indicated that the overall model was statistically
significant, χ2(11, N=12,984) = 3985.495, p<.001. Furthermore, data indicated that 77.8% of
cases were categorized correctly. In terms of individual predictors, results of demographic
variables, χ2(4, N=12,984)= 2173.265, p<.001, showed that older adolescents were almost 1.4
times (OR=1.390, 95% CI=1.339-1.443) more likely to use alcohol than younger ones; African
American adolescents were 1.36 times (OR=.736, 95% CI=.647-.837) less likely to drink than
white youths; Asian American adolescents were 2 times (OR=.500, 95% CI=.388-.644) less
likely to use alcohol than white counterparts; there was no significant difference in alcohol use
between female and male adolescents in the study. Findings of parental set of variables, χ2(3,
N=12,984) = 825.401, p<.001, indicated that these who received high parental support in study
was 1.39 times (OR=.719, 95% CI=.651-.795) less likely to use alcohol than those who received
low parental support; similarly, those who received strong parental disapproval were almost 2.3
times (OR=.436, 95% CI=.385-.495) less likely to drink alcohol than everybody else; however,
there was no significant difference in alcohol drinking between “low parental monitoring” and
“high parental monitoring” participants. The third block, χ2(2, N=12,984) = 938.619, p<.001,
reported that the more substance-using friends the more likely the participants used alcohol
(OR=1.355, 95% CI=1.314-1.397); those who received strong peer disapproval were 2.65 times
(OR=.387, 95% CI=.348-.431) less likely to drink alcohol than everyone else. The independent
variable - religiosity, χ2(1) = 35.172, p<.001, revealed that the more religious the less likely
59
(1.22 times) the adolescents use alcohol (OR=.819, 95% CI=.767-.875). Result of the school-
based prevention program model indicated that those who attended a special class on drugs or
alcohol were 1.2 times (OR=.831, 95% CI=.752-.919) less likely to use alcohol than those who
did not attend the prevention training programs, χ2(1, N=12,984) = 13.136, p<.001.
Table 9. Binary Logistic Regression Analysis Examining Alcohol Use (N=12,984)
Variables
B (SE) Wald (X2) OR (95% CI)
Block 1
Age
African American
Asian American
Gender (1)
.329 (.019)
-.307 (.066)
-.693 (.129)
.064 (.051)
299.206***
21.634***
28.876***
1.628
1.390 (1.339-1.443)
.736 (.647-.837)
.500 (.388-.644)
1.067 (.966-1.178)
Block 2
Parental monitoring (1)
Parental support (1)
Parental disapproval (1)
-.092 (.051)
-.329 (.051)
-.829 (.064)
3.249
41.644***
165.765***
.912 (.826-1.008)
.719 (.651-.795)
.436 (.385-.495)
Block 3
Peer substance use
Peer disapproval (1)
.304 (.016)
-.948 (.054)
378.148***
304.597***
1.355 (1.314-1.397)
.387 (.348-.431)
Block 4
Religiosity
-.200 (.034)
35.172***
.819 (.767-.875)
Block 5
Prevention programs
-.185 (.051)
13.125***
.831 (.752-.919)
Note: White is the reference group for race.
For Model: R2 = .303 (Cox & Snell), R2 = .420 (Nagelkerke), χ2(11) = 3985.495, p<.001
Block 1: R2 = .178 (Cox & Snell), R2 = .247 (Nagelkerke), χ2(4)= 2173.265, p<.001
Block 2: R2 = .238 (Cox & Snell), R2 = .329 (Nagelkerke), χ2(3) = 825.401, p<.001
Block 3: R2 = .300 (Cox & Snell), R2 = .416 (Nagelkerke), χ2(2) = 938.619, p<.001
Block 4: R2 = .302 (Cox & Snell), R2 = .419 (Nagelkerke), χ2(1) = 35.073, p<.001
Block 5: R2 = .303 (Cox & Snell), R2 = .420 (Nagelkerke), χ2(1) = 13.136, p<.001
*p<.05, **p<.01, ***p<.001.
Summary: Results of bivariate and multivariate analyses support hypotheses of the study. The
inconsistent findings about Parental monitoring in bivariate and multivariate analyses would be
due to redundancy of the variable, presumably, with other parenting measures. Specifically,
parental monitoring wasn't significant in multivariate analyses because of shared variance with
all those other parenting variables.
60
5.5 MODERATION AND MEDIATION TESTS ON THE IMPACTS OF RACE, AGE,
GENDER, AND RELIGIOSITY ON MARIJUANA AND ALCOHOL USE.
To further investigate past research findings about the relationships among demographic
variables, religiosity, and adolescent substance use, this section focuses on moderation and
mediation analyses to explore (1) if the relationships between religiosity and adolescent alcohol
and marijuana use are moderated by age, race, and gender (Figure 2) , and (2) if religiosity acts
as a mediator in the relationships between background variables and adolescent substance use
variables as addressed by research questions 2 and 3 as well as purposes 4 and 5 in this study
(Figure 3).
Marijuana/
Alcohol
Use
Gender/Age/
Race
Religiosity
Figure 2. Gender as a moderator for the impact of religiosity on alcohol and marijuana use
Figure 3. Religiosity as a mediator for the impact of age, race, and gender on alcohol and marijuana use
Religiosity
(Z)
Alcohol &
Marijuana use (Ys)
Age, Race,
Gender (Xs)
61
5.5.1 Race, age, and gender as moderators for the relationships between religiosity and
alcohol and marijuana use
Results of moderation tests revealed that there were no moderation effects of age and race in the
relationship between religiosity and alcohol use; similarly race and age did not serve as
moderators of the impact of religiosity on marijuana use among the study participants. However,
gender did serve as a moderator for the relationship between religiosity and marijuana use among
the adolescents. The moderation effect of gender on the impact of religiosity on marijuana was
evaluated by the interaction term which indicated that the lower likelihood of marijuana use by
religious youth was especially apparent for girls (B= -.17, Wald chi square= 4.31, p< .05). The
same pattern was found for alcohol use but the effect was not statistically significant (p< .15).
Whereas age and race did not qualify the religiosityuse relationship, female gender did
contribute to a stronger impact of religiosity on lower usage.
5.5.2 Religiosity as a mediator of race, age, and gender predicting marijuana and alcohol
use
Findings from previous research and the bivariate correlations in this study revealed a potential
mediation effect of religiosity (Z) on the relationships of age, race, and gender (Xs) with
marijuana and alcohol use (Ys) among the study participants. Thus, this section will check: (1)
the relationship between independent variables of race, age, and gender (Xs) and religiosity (Z),
controlling for the parenting and peer variables, and (2) the relationships between religiosity and
the Ys. Specifically, I will run an OLS regression analysis predicting religiosity (Z) from age,
race, gender, parenting, and peer variables (Xs). Then, I will also check the relationships
between Xs and Ys in binary logistic regression analyses with and without the presence of
religiosity in the models. These latter analyses provide the estimates for the effect of religiosity
62
(ZY) and show possible differences in the X--Y prediction when religiosity is absent versus
present in the analysis (Babsent/Bpresent). In the mediation figures presented later, the predictors of
religiosity are non-standardized Bs from the OLS regression, and the estimates of the Ys are
logistic regression coefficients from the binary logistic regression analyses.
5.5.2.1 Tests of the Mediating Role of Religiosity in the Age, Gender and Race effects on
alcohol and marijuana use
Figure 4 shows the ordinary regression B and logistic regression coefficients from the analyses
predicting religiosity and marijuana use, respectively. Religiosity was a strong predictor of less
marijuana use (B= -.398, p< .001). This figure also provides the XZ coefficients, revealing
that Black race (B= .187) and female gender (B=.106) significantly predicted religiosity (Z).
These results informally supported an indirect path in which black race and female gender
enhance religiosity, which, in turn, reduces the likelihood of marijuana use. The direct logistic B
coefficients for the background predictors showed that younger age, Asian ethnicity and female
gender were associated with lower likelihood of marijuana use, but Black race was associated
with higher likelihood. Only Black race and female gender are interesting in the mediation
context, since they are significantly related to religiosity. The logistic coefficients became more
positive (for Black race) and less negative (for gender) when religiosity was in the model as a
mediator that reflects the indirect impact that contributes to lower usage.
Figure 5 shows the same mediation model using alcohol use as the outcome (Y) variable.
Once again religiosity is a strong direct predictor of Y, alcohol use (B= -.197). So the mediation
paths from Black race and female gender to greater religiosity and from religiosity to lower use,
were supported. Regarding alcohol use, Black race, younger age and Asian ethnicity were
associated with lower usage, but gender was unrelated to likelihood of alcohol use (although
63
female gender indirectly reduced alcohol). Once again the logistic coefficients for the
background predictors of Black race and gender were more positive when religiosity was
controlled due to the inclusion of the negatively signed indirect effect through religiosity. Part of
the initial Black race and female gender effect is due to drug use diminishing influence of
religiosity. When that part is removed from Y by entering/controlling religiosity, the
relationships of the predictors to Y are more positive (or less negative). In the absence of the
control for religiosity, Black race and female gender were over-estimated as explainers of
alcohol and marijuana usage and marijuana use diminishing contribution of religiosity was lost.
In general the mediation effects found and reported here are examples of partial
mediation since they occur in the presence of significant direct effects of the background
variables on the outcome usage variables.
64
Figure 4. Religiosity as a partial mediator for African Americans and females in marijuana use (N=12,984)
Figure 5. Religiosity as a partial mediator for African Americans and females in alcohol use (N=12,984)
B(Age)=.004
B(Asian)= .001
B(Black)= .187***
B(Female)=.106***
B(Age)= .320***/.321***
B(Asian)= -.389*/-.348*
B(Black)=.265***/.388***
B(Female)=-.192**/-.162**
Religiosity
(Z)
B= -.398***
Marijuana
use (Y) Age, Race,
Gender (Xs)
B(Age)=.004
B(Asian)= .001
B(Black)= .187***
B(Female)=.106***
B(Age)= .332***/.337***
B(Asian)= -.699***/-.701***
B(Black)=-335***/-.293***
B(Female)=.035/.055
Religiosity
(Z)
B= -.197***
Alcohol use
(Y) Age, Race,
Gender Xs)
65
6.0 DISCUSSION
6.1 DISCUSSION OF SIGNIFICANT FINDINGS
6.1.1 Findings of main analyses predicting marijuana and alcohol use
Findings of this study confirm the proposed hypotheses that religiosity, school-based prevention
programs, parental support, parental monitoring, parental disapproval, peer use, and peer
disapproval all together significantly predict alcohol and marijuana use among the study
participants. Additionally, higher religiosity, attending alcohol and drug training programs,
higher parental support, parental disapproval, peer disapproval, and less peer use are related to
lower likelihood of marijuana and alcohol use, controlling for background factors.
This study overcomes shortcoming of previous studies on religiosity by using the most
recent national data set, five items of individual religiosity, sum of Z scores of religiosity
measures, and importantly checking reliability of the religious measures. Therefore, these
findings do provide additional evidence of the association between religiosity and marijuana and
alcohol use, thus reinforcing the extant research on religiosity. These findings suggest that
religiosity can be used as a protective factor to help adolescents deal with alcohol and marijuana
problems. Social workers are trained to empower disadvantaged and vulnerable population. They
are also equipped with knowledge and skills to work with individuals and families. Therefore,
social workers are the right people who can effectively help adolescents avoid using alcohol and
marijuana or maintain their sobriety by using individual religiosity as a prevention and treatment
method. Using religiosity as a prevention and treatment method for adolescents is needed and
appropriate for adolescents since they are still in developmental stages. Thus, rehabilitation
works better than punishment for them. Punishment such as incarceration may result in more
problem behaviors among adolescents, because they are isolated from society and they can easily
66
have bad influences from other peers in incarceration settings. Besides, using religiosity as a
rehabilitation approach reflects humanity in drug policy and social work practice. Findings about
religiosity in this study are also consistent with results of the review by Johnson, De Li, Larson,
and McCullough (2000) who concluded that studies that used four or more religious measures
consistently reported beneficial effects of religiosity on substance use.
Given the effectiveness of school-based prevention programs and their cost-efficiency in
comparison with incarceration, these findings suggest that implementation of prevention
programs for adolescents at schools is necessary. With their core values such as collaboration
and therapeutic alliance, social workers can work with schools and families and take the lead in
implementing school-based prevention programs for adolescents. These programs can help
adolescents avoid using substances or change their problem behaviors, meanwhile, adolescents
can still receive support from their families and friends. Using prevention programs for
adolescents to replace incarceration can also help Federal and States governments save their
annual budgets for law enforcement, which is much more costly than prevention and treatment
methods. Since there are still controversial findings about the effectiveness of school-based
prevention programs, these findings do provide additional discoveries in the extant research in
several perspectives. The finding of beneficial effects of school-based prevention programs on
marijuana use in this study supports and is consistent with previous finding in a longitudinal
study by Botvin, Baker, Dusenbury, Tortu, & Botvin (1990). With regards to alcohol use, results
of two previous longitudinal studies by Hansen, Malotte, & Fielding (1988) and Botvin, Baker,
Dusenbury, Tortu, & Botvin (1990) indicated that school-based prevention programs failed to
prevent adolescents from drinking alcohol. Conversely, this study does confirm the beneficial
effects of school-based prevention programs on alcohol use among the study participants.
67
Furthermore, literature review shows that none of the extant studies have, so far, examined the
impact of school-based prevention programs together with a variety of demographic, personal
and environmental variables like in this study. Therefore, this finding can make a significant
contribution to the current research on the impacts of school-based prevention programs on
adolescent substance use.
6.1.2 Findings of moderation and mediation tests
Similar to previous findings (Wills, Yaeger, & Sandy, 2003; Pitel et al., 2012), this study also
confirms that the moderation effects on lower use of alcohol and marijuana are stronger among
female gender, which make an added contribution the extant research.
Results from Tables 3, 8, 9, and the OLS regression analyses demonstrated that
religiosity was not impactful among Asian American adolescents; and Asian American youths
were much less likely to drink alcohol and use marijuana than white and African American
counterparts irrespective of their religious beliefs. The later confirms the previous finding by
Barnes, Welte, and Hoffman (2002) that Asian American adolescents have the lowest level of
alcohol and drug use in comparison with Whites, Blacks, Hispanics, and other races in the U.S.
However, the former needs further examination in the future studies since religious measures in
this study strongly focused on church and church activities, which leave out common religions
among Asian population such as Buddhism and Hinduism.
Results of exploratory tests (Figures 4 and 5) indicated that African American youth and
female adolescents have greater religiosity, which supports the current research findings (Brown,
Parks, Zimmerman, & Phillips, 2001; and Rote & Starks, 2010; Wallace, Brown, Bachman, &
LaVeist, 2003). Similarly, religiosity significantly predicted alcohol and marijuana use among
the study participants, which supports the existing studies by Jang & Johnson (2001); Vaughan,
68
de Dios, Steinfeldt, and Kratz (2011); Johnson, Larson, and McCullough (2000). Besides, the
unstandardized coefficients of African American youth and female gender predicting alcohol and
marijuana use significantly changed with the presence of religiosity in the model. These three
elements confirm the partial mediation effects of religiosity on alcohol and marijuana use among
African American youth and female adolescents. Specifically, greater religiosity among African
American youth and female adolescents indirectly reduces their likelihood of drinking and using
marijuana. Based on this critical finding, social workers can focus on increasing individual
religiosity among African American youth and female adolescents and use it to protect them
from using the substances. This can be done in numerous ways. For instance, social workers can
work with families and churches to get African American youth and female adolescents involved
in church activities such as religious singing and dancing, or encourage them to regularly attend
church services, or establish bible-study groups for them. These strategies can help African
American youth and female adolescents increase their individual religiosity, which consequently
reduce their substance use.
6.2 LIMITATION OF THE STUDY
This study certainly has several limitations. Because it is cross-sectional, the current study could
not take into account change in marijuana and alcohol use over time among the study
participants. Additionally, measures of dependent variables and school-based prevention
programs variable in this study were binary and crude that I did not have or use a more refined
use measure. Besides, religious measures in this study are not strongly related to religions of
Asian population, which consequently affects the study findings about Asian American
adolescents. Moreover, types of religions were not specified in the study, which limits our
understanding of the potentially differential impact of specific religions. Furthermore, types and
69
methods of school-based prevention programs were not clearly addressed in this study, which
also limits our understanding of the impact of the prevention programs.
6.3 IMPLICATIONS
Despite its limitations, this study provides several implications for social work practice, future
research on adolescent substance use, and drug policy. For social work practice, results of this
study suggest that religious beliefs, parental support, parental disapproval, peer disapproval, and
substance-using friends are influential factors to alcohol and marijuana use among the study
participants. Besides, implementation of school-based prevention programs for these adolescents
is extremely needed to prevent them from using alcohol and marijuana. Given their numerous
strengths in working with individuals and families, social workers can effectively combine these
personal and environmental factors to help white, African American, and Asian American
adolescents and their families deal with alcohol and marijuana problems. Specifically, these
findings suggest that social workers should implement school-based prevention programs which
provide the adolescents with skills and knowledge to deal with alcohol and marijuana problems.
For instance, the programs should provide white, African American and Asian American
adolescents with skills to deal with substance-using friends such as how to refuse or avoid using
substances when they are offered by their peers; and how to wisely confront their peers’
substance use when needed. In addition, social workers should collaborate with families, schools
and churches to design programs or training sessions to increase and strengthen individual
religiosity among the adolescents. These programs need to make the adolescents recognize the
importance of religiosity in dealing with substance use. In tandem with that, they need to
encourage the adolescents to act in accordance with their religious beliefs and link negative
consequences of alcohol and marijuana use with their individual religiosity. It is recommended
70
that contents and activities of the programs should be designed to strongly focus on factors
related to cultures and religions of white, African Americans and Asian Americans. For example,
programs for Asian American adolescents should be based on Buddhism or Hinduism’s
philosophy and beliefs, depending on their religions. In line with increasing and strengthening
individual religiosity for the adolescents, social workers should closely work with parents and
peers to encourage their strong support, monitoring, and disapproval towards substance-using
behaviors among the adolescents. To help parents and peers effectively fulfill their supportive
role, social workers should provide them with skills to (1) establish close and intimate
relationships with the adolescents, which helps them identify substance-using behaviors and to
(2) effectively deal with substance-using behaviors among the adolescents. It is uneasy to get
families and adolescents involved in such training programs sometimes due to numerous barriers
such as transportation and child care. Therefore, social workers should use different strategies to
stimulate active participation of families and adolescents in the training programs such as
providing transportation tickets, child care, gift vouchers, and raffle tickets. Combining these
methods, social workers could potentially make a significant contribution to lessening marijuana
and alcohol use problems among white, African American, and Asian American youths.
With regards to contribution to future research, findings of the exploratory analyses
suggest some potentially important discoveries which require more extensive research in the
future. One of the important discoveries is the partial mediation effect of religiosity on substance
use among African American adolescents. These finding indicates that their strong religious
beliefs partially ameliorate substance use among African American youth. The partial mediation
effect of religiosity on alcohol use among female adolescents also needs further investigations;
71
their likelihood of drinking significantly increased with the presence of religiosity in the model
despite the fact that they have greater religiosity which should reduce their drinking likelihood.
Findings of this study also suggest some implications for policy makers. Under social
work perspective, using prevention and treatment methods to help adolescents deal with alcohol
and marijuana problems is strongly encouraged to replace current law enforcement strategy
because of humanity and effectiveness of these methods. Incarcerating adolescents who have
alcohol and marijuana problems is unnecessary since it is not effective, and importantly it does
not reflect humanity in drug policy. These adolescents can change their problem behaviors with
active support from social workers, schools, families, and friends through implementation of
combined programs for the adolescents, their families, and friends. At a macro level, drug policy
plays a very important role in ensuring effectiveness of these programs. Presently, lack of
funding, workforce development, and inadequate compensation for service workers are major
barriers for the implementation of substance use programs for adolescents (Cavanaugh, Kraft,
Muck, & Merrigan, 2011). Therefore, it is suggested that drug policy should cut down budget for
incarceration and allocate adequate funds to substance use programs for adolescents. These
efforts will help expand and improve current services in substance use programs for adolescents,
giving them more opportunities to access to services they need. Besides, drug policy should
facilitate professional development and capacity building for social workers who work with
substance-using adolescents. For example, drug policy should encourage states to establish
cross-training programs for social workers to strengthen their knowledge and skills to work with
adolescents. Such trainings are essential for social workers since current substance use programs
for adolescents are based on adults’ models, and service providers are lack of expertise in
working with adolescent population (Cavanaugh & White, 2003). In line with capacity building,
72
it is essential to hire qualified social workers who can ensure effectiveness of substance use
programs for adolescents. Currently, low-paid job and stressful working environment are major
bariers for recruiting and retaining qualified social workers in substance use programs. Thus,
drug policy should have adequate compensation for social workers who work with substance-
using clients. These changes in drug policy will make a significant contribution to improving
effectiveness of substane use programs for adolescents, which consequently lessen current
alcohol and marijuana problems in the United States.
73
BIBLIOGRAPHY
Acker, C.J. (1993). Stigma or legitimation? A historical examination of the social potential of
addiction disease models. Journal of Psychoactive Drugs, 25, 193-205.
Albrecht, S. L., Chadwick, B. A., and Alcom, D. (1977). Religiosity and deviance: Application of
an attitude-behavior contingent consistency model. Journal for the Scientific Study of
Religion, 16, 263-274.
Ali, M. M., & Dwyer, D. S. (2010). Social network effects in alcohol consumption among
adolescents. Addictive Behaviors. 35(4), 337-342.
Averna, S. & Hesselbrock, V. (2001). The relationship of perceived social support to substance
use in offspring of alcoholics. Addictive Behaviors, 26, 363–374.
Bahr, S. J., & Hoffmann, J. P. (2008). Religiosity, peers, and adolescent drug use. Journal of
Drug Issues, 38(3), 743-769.
Bahr, S. J., Hawks, R. D., & Wang, G. (1993). Family and religious influences on adolescent
substance abuse. Youth and Society, 24, 443-465.
Bahr, S. J., Hoffmann, J.P., and Yang, X. (2005). Parental and peer influences on the risk of
adolescent drug use. The Journal of Primary Prevention, 26(6), 529- 551.
Bahr, S.J., and Hoffmann, J.P. (2010). "Parenting style, religiosity, peers, and adolescent heavy
drinking." Journal of Studies on Alcohol and Drugs, 539-543.
Baier, C.J., and Wright, B.R.E. (2001). “If you love me keep my commandments: A meta-
analysis of the effect of religion on crime.” Journal of Research in Crime and
Delinquency, 38, 3–21.
Bandura, A. (1977). Social learning theory. New York, NY: General Learning Press.
Barnes, G.M., Reifman, A.S., Farrell, M.P., & Dintcheff, B.A. (2000). The effects of parenting
on the development of adolescent alcohol misuse: A six-wave latent growth model.
Journal of Marriage and the Family, 62, 175-186.
Barnes, G.M., Welte, J.W., and Hoffman, J.H. (2002). Relationship of alcohol use to delinquency
and illicit drug use in adolescents: Gender, age, and racial/ethnic differences. Journal of
Drug, 32(1), 153-178.
Barnett, E., et al. (2012). Motivational Interviewing for adolescent substance use: A review of
the literature. Addictive Behaviors, 37(12), 1325-1334.
Blum, R.W., Beuhring, T., Shew, M.L., Bearinger, L.H., Sieving, R.E., and Resnick, M.D. (2000).
The effects of race/ethnicity, income, and family structure on adolescent risk behaviors.
American Journal of Public Health, 90(12), 1879–1884.
Burkett, S. R, and White, M. (1974). Hellfire and delinquency: Another look. Journal for the
Scientific Study of Religion,13, 455-462.
Branstetter, S. A., Low, S., & Furman, W. (2011 ). The influence of parents and friends on
adolescent substance use: A multidimensional approach. Journal of Substance Use,
16(2), 150-160.
Braucht, G.N., Follingstad, D., Brakarsh, D., and Berry, K.L. (1973). Drug education: A review of
goals, approaches and effectiveness, and a paradigm for evaluation. Quarterly Journal of
Studies on Alcohol, 34, 1279-1292.
Bray, J. H., Adams, G. J., Getz, J. G., & McQueen, A. (2003). Individuation, peers, and
adolescent alcohol use: A latent growth analysis. Journal of Consulting and Clinical
Psychology, 71(3), 553-564.
74
Brook, J. S., Brook, D. W., Zhang, C., & Cohen, P. (2009). Pathways from Adolescent Parent-
Child Conflict to Substance Use Disorders in the Fourth Decade of Life. American
Journal On Addictions, 18(3), 235-242.
Brown, T. L., Parks, G. S., Phillips, C. M., & Zimmerman, R. S. (2001). The role of religion in
predicting adolescent alcohol use and problem drinking. Journal of Studies on Alcohol,
62(5), 696-705.
Catalano, R. F., Morrison, D. M., Wells, E. A., Gillmore, M. R., Iritani, B., & Hawkins, J. D.
(1992). Ethnic differences in family factors related to early drug initiation. Journal of
Studies on Alcohol, 53, 208–217.
Cavanaugh, D. A., & White, A. (2003). Adolescent substance abuse treatment system summit
report. Unpublished report submitted to the Center for Substance Abuse Treatment,
SAMHSA and the Robert Wood Johnson Foundation.
Cavanaugh, D., Kraft, M.K., Muck, R., & Merrigan, D.M. (2011). Toward an effective treatment
system for adolescents with substance use disorders: The role of the states. Children and
Youth Services Review. 33, Supplement 1(0), S16-S22.
Centers for Disease Control and Prevention. (2010). Alcohol and public health, fact sheets,
underage drinking. Retrieved May 16, 2013, from http://www.cdc.gov/alcohol/fact-
sheets/underage-drinking.htm
Chadwick, B. A., & Top, B. L. (1993). Religiosity and delinquency among LDS adolescents.
Journal for the Scientific Study of Religion, 32, 51-67.
Chaplin, T.M., Sinha, R., Simmons, J.A., Healy, S.M., Mayes, L.C., Hommer, R.E., & Crowley,
M.J. (2012). Parent–adolescent conflict interactions and adolescent alcohol use. Addictive
Behaviors, 37(5), 605-612.
Chatterji, P. (2006). Illicit drug use and educational attainment, Health Economics.15 (5), 489-
511.
Chung, H. J. (1997). Religiosity and substance use among Asian Americans. (Order No. 1387574,
California State University, Long Beach). ProQuest Dissertations and Theses, 89-89.
Corwyn, R.F., and Benda, B.B. (2002). The Relationship between use of alcohol, other drugs, and
crime among adolescents: An Argument for a Delinquency Syndrome. Alcoholism
Treatment Quarterly, 20(2), 35-49.
Curran, P. J., Stice, E., & Chassin, L. (1997). The relation between adolescent alcohol use and
peer alcohol use: A longitudinal random coefficients model. Journal of Consulting and
Clinical Psychology, 65(1), 130-140.
De Leo, J. A., & Wulfert, E. (2013). Problematic internet use and other risky behaviors in
college students: An application of problem-behavior theory. Psychology of Addictive
Behaviors, 27(1), 133-141.
DiClemente, R. J., Wingood, G. M., Crosby, R., Sionean, C., Cobb, B. K., Harrington, K.,
Davies, S., Hook III, E. W., and Kim, M. O. (2001). Parental monitoring: Association
with adolescents risk behaviours. Pediatrics, 107, 1363-1368.
Dishion, T.J., & Loeber, R. (1985). Adolescent marijuana and alcohol use: The role of parents
and peers revisited. American Journal of Drug and Alcohol Abuse, 11, 11-25.
Donahue, M., and Benson, P. (1995). Religion and the well-being of adolescents. Journal of
Social Issues, 51, 145-160.
Donovan, J. E. (2004). Adolescent alcohol initiation: A review of psychosocial risk factors.
Journal of Adolescent Health, 35, 529.e7– e18.
75
Drapela, L. A., and Mosher, C. (2007). The conditional effect of parental drug use on parental
attachment and adolescent drug use: Social control and social development model
perspectives. Journal of Child & Adolescent Substance Abuse, 16(3), 63-87.
Durrant, R and Thakker, J. (2003). Substance use and abuse: Cultural and historical perspectives.
California: California: Thousand Oaks.
Epstein, J. A., Botvin, G. J., Baker, E., & Diaz, T. (1999). Impact of social influences and
problem behavior on alcohol use among inner-city Hispanic and black adolescents.
Journal of Studies on Alcohol, 60, 595–604.
Erickson, C. K. (2007). The science of addiction: From neurobiology to treatment. New York:
W.W. Norton.
Erickson, K. G., Crosnoe, R., & Dornbusch, S. M. (2000). A social process model of adolescent
deviance: Combining social control and differential perspectives. Journal of Youth and
Adolescence, 29, 395–425.
Farrell, A. D. (1994). Structural equation modeling with longitudinal data: Strategies for
examining group differences and reciprocal relationships. Journal of Consulting and
Clinical Psychology, 62, 477- 487.
Farrell, A. D., & White, K. S. (1998). Peer influences and drug use among urban adolescents:
Family structure and parent-adolescent relationship as protective factors. Journal of
Consulting & Clinical Psychology, 66(2), 248-258.
Farrell, A., & Danish, S. (1993). Peer drug associations and emotional restraint: Causes or
consequences of adolescents' drug use? Journal of Consulting and Clinical Psychology,
43, 522-527.
Foley, K. L., Altman, D., Durant, R. H., & Wolfson, M. (2004). Adults’ approval and
adolescents’ alcohol use. Journal of Adolescent Health, 35, 345.e17–26.
Fraser, M. (1984). Family, school, and peer correlates of adolescent drug abuse. Social Service
Review, 58, 434-447.
Gahlinger, P. M. (2001). Illegal drugs: A complete guide to their history, chemistry, use, and
abuse. Salt Lack, Utah: Sagebrush Press.
Goncy, E. A., & van Dulmen, M.,H.M. (2010). Father do make a difference: Parental
involvement and adolescent alcohol use. Fathering, 8(1), 93-108.
Goodman, E., and Huang, B. (2002). Socioeconomic status, depressive symptoms and adolescent
substance abuse. Archives of Pediatric and Adolescent Medicine, 156, 448-453.
Hanson, M.D., and Chen, E. (2007). Socioeconomic status and substance use behaviors in
adolescents: the role of family resources versus family social status, Journal of health
Psychology. 12(1), 32-35.
Hawkins, J. D., & Weis, J. G. (1985). The social developmental model: An integrated approach
to delinquency prevention. Journal of Primary Prevention, 6, 73–97.
Heath, D. B. (1989). The new temperance movement: Through the looking glass. Drug and
Society, 3, 143-168.
HeavyRunner-Rioux, A. R., & Hollist, D. R. (2010). Community, family, and peer influences on
alcohol, marijuana, and illicit drug use among a sample of Native American youth: An
analysis of predictive factors. Journal of Ethnicity in Substance Abuse, 9(4), 260-283.
Higgins, P. C., and Albrecht, G. L. (1977). Hellfire and delinquency revisited. Social Forces, 55,
952-958.
Hirschi, T. (1969). Causes of delinquency. Berkeley, CA: University of California Press.
Hirschi, T., and Stark, R. (1969). “Hellfire and Delinquency.” Social Problems, 17, 202–213.
76
Hoffmann, J. P., & Johnson, R. A. (1998). A national portrait of family structure and adolescent
drug use. Journal of Marriage and the Family, 60, 633-645.
Hood, R., Jr., Spilka, B., Hunsberger, B.,& Gorsuch, B. (1996). The psychology of religion (2nd
ed.). New York: Guilford.
Huang, G.C., Unger, J.B., Soto, D., Fujimoto, K., Pentz, M.A., Jordan-Marsh, M., & Valente,
T.W. (2014). Peer Influences: The Impact of Online and Offline Friendship Networks on
Adolescent Smoking and Alcohol Use. Journal of Adolescent Health, 54(5), 508-514.
Humensky, J.L. (2010). Are adolescents with high socioeconomic status more likely to engage in
alcohol and illicit drug use in early adulthood? Substance Abuse Treatment, Prevention,
and Policy, 5(19), 1-10.
Humphreys, K & McLellan, A. T. (2010). Brief intervention, treatment, and recovery support
services for Americans who have substance use disorders: An overview of policy in the
Obama administration. Psychological Services, 7(4), 275-284.
Jang, S. J. and Johnson, B. (2001). “Neighborhood disorder, individual religiosity, and
adolescent drug use: A test of multilevel hypotheses.” Criminology, 39, 501–35.
Jessor, R. (1976). Predicting time of onset of marijuana use: A developmental study of high
school youth. Journal of Consulting and Clinical Psyclwlogy, 44, 125-134.
Jessor, R. (1987). Problem-behavior theory, psychosocial development, and adolescent problem
drinking. British Journal of Addiction, 82(4), 435-446.
Jessor, R., & Jessor, S. L. (1977). Problem behavior and psychosocial development: A
longitudinal study of youth. New York: Academic Press.
Jessor, R., Chase, J. A., & Donovan, J. E. (1980). Psychosocial correlates of marijuana use and
problem drinking in a national sample of adolescents. American Journal of Public
Health, 70, 604-613.
Jessor, R., Turbin, M. S., Costa, F. M., Dong, Q., Zhang, H., and Wang, C. (2003). Adolescent
problem behavior in China and the United States: A cross-national study of psychosocial
protective factors. Journal of Research on Adolescence, 13(1), 329–360.
Johnson, B., De Li, S., Larson, D., and McCullough, M. (2000). “A systematic review of the
religiosity and delinquency literature: A research note.” Journal of Contemporary
Criminal Justice, 16, 32–52.
Johnston, L. D., O’Malley, P. M., & Bachman, J. G. (2001). Monitoring the Future national
survey results on drug use, 1975–2000. Vol. I. Secondary school students (NIH
Publication No. 01–4924). Bethesda, MD: National Institute on Drug Abuse.
Kelly, A. B., O’Flaherty, M., Toumbourou, J. W., Connon, J. P., Hemphill, S. A., & Catalano, R.
F. (2011). Gender differences in the impact of families on alcohol use: A lagged
longitudinal study of early adolescents. Addiction, 106, 1427–1436.
Kelly, J.F., Pagano, M.E., Stout, R.L., Johnson, S.M. (2011). Influence of religiosity on 12-Step
participation and treatment response among substance-dependent adolescents. Journal of
Studies on Alcohol and Drugs, 72(6), 1000-1011.
Kleiman, M. A., & Hawdon, J. E. (2011). Encyclopedia of drug policy. California, CA: SAGE.
Korsmeyer, P., and Kranzler, H. R. (2009). Opium: U.S. Overview. Encyclopedia of Drugs,
Alcohol & Addictive Behavior, 3, 183-190.
Levinson, M. H. (2002). The drug problem: a new view using the general semantics approach.
Westport, CT: Praeger.
77
Lewis, M. L., and Jordan, L. C. (2005). Paternal Relationship Quality as a Protective Factor:
Preventing Alcohol Use Among African American Adolescents. Journal of Black
Psychology, 31(2), 152-171.
Lo, C.C and Cheng, T. C. (2011). Racial/Ethnic Differences in Access to Substance Abuse
Treatment. Journal of Health Care for the Poor and Underserved. 22(2), 621-637.
Low, S., Shortt, J. W., and Snyder, J. (2012). Sibling influences on adolescent substance use:
The role of modeling, collusion, and conflict. Development and Psychopathology, 24(1),
287-300.
Mahon-Halt, T., and Mosher, C. (2011). D.A.R.E. In Kleiman, M. A., & Hawdon, J. E. (Eds).
Encyclopedia of drug policy. (pp.187-189). California, CA: SAGE.
Malhotra, A., and Biswas, P. (2006). Cannabis Use and Performance in Adolescents. Journal of
Indian Association for Child and Adolescent Mental Health, 2(2), 59-67.
Marcos, A.C., Bahr, S.J., & Johnson, R.C. (1986). Test of a bonding/association theory of
adolescent drug use. Social Forces, 65, 135-161.
Marshal, M. P., & Chassin, L. (2000). Peer Influence on Adolescent Alcohol Use: The
Moderating Role of Parental Support and Discipline. Applied Developmental Science,
4(2), 80-88.
Martino, S. C., Ellickson, P. L., & McCaffrey, D. F. (2009). Multiple trajectories of peer and
parental influence and their association with the development of adolescent heavy
drinking. Addictive Behaviors, 34, 693–700.
Mason, M., Mennis, J., Linker, J., Bares, C., & Zaharakis, N. (2013). Peer Attitudes Effects on
Adolescent Substance Use: The Moderating Role of Race and Gender, Prevention Science,
1-9.
Mason, M., Mennis, J., Linker, J., Bares, C., & Zaharakis, N. (2014). Peer Attitudes Effects on
Adolescent Substance Use: The Moderating Role of Race and Gender. Prevention
Science, 15(1), 56-64.
Maxwell, K. A. (2002). Friends: The role of peer influence across adolescent risk behaviors.
Journal of Youth and Adolescence, 31(4), 267-277.
McFarling, L., D'Angelo, M., Drain., Marsha., Gibbs, D. A., Rae O., and Kristine, L. (2011).
Stigma as a barrier to substance abuse and mental health treatment. Military Psychology,
23(1), 1-5.
Miller, H., Jennings, W., Alvarez-Rivera, L., & Miller, J. (2008). Explaining substance use
among Puerto Rican adolescents: A partial test of Social learning theory. Journal of Drug
Issues, 38(1), 261-283.
Miller, J.W., Naimi, T.S., Brewer, R.D., and Jones, S.E. (2007). Binge drinking and associated
health risk behaviors among high school students. Pediatrics, 119, 76–85.
Mobley, M., & Chun, H. (2013). Testing Jessor's problem behavior theory and syndrome: A
nationally representative comparative sample of Latino and African American
adolescents. Cultural Diversity and Ethnic Minority Psychology, 19(2), 190-199.
Moon, S.S., Blakey, J.M., Boyas, J., Horton, K., & Kim, Y.J. (2014). The Influence of Parental,
Peer, and School Factors on Marijuana Use Among Native American Adolescents.
Journal of Social Service Research, 40(2), 147-159.
Morgan, H. W. (1981). Drugs in America: A social history, 1800-1980. Syracuse: Syracuse
University Press.
78
Mrug, S., & McCay, R. (2013). Parental and peer disapproval of alcohol use and its relationship
to adolescent drinking: Age, gender, and racial differences. Psychology of Addictive
Behaviors, 27(3), 604-614.
Nash, S.G., McQueen, A., & Bray, J.H. (2005). Pathways to adolescent alcohol use: family
environment, peer influence, and parental expectations. Journal of Adolescent Health,
37(1), 19-28.
National Institute on Drug Abuse. (2010). Drug, Brain, and Behaviors: the Science of Addiction.
Retrieved January 3, 2013, from
http://www.drugabuse.gov/sites/default/files/sciofaddiction.pdf
National Institute on Drug Abuse. (2012). Principles of Drug Addiction Treatment: A Research-
Based Guide (3rd Edition. Retrieved January 3, 2013, from
http://www.drugabuse.gov/sites/default/files/podat_1.pdf
Ndugwa, R. P., Kabiru, C. W., Cleland, J., Beguy, D., Egondi, T., Zulu, E. M., and Jessor, R.
(2011). Adolescent problem behavior in Nairobi’s informal settlements: applying
problem behavior theory in sub-Saharan Africa. Journal of Urban Health. 14 (Suppl
2):S298–S317.
Office of the National Drug Control Policy. (2014). The National Drug Control Strategy: A 21st
Century Approach to Drug Policy. Retrieved July 14, 2014 from
http://www.whitehouse.gov/sites/default/files/ondcp/policy-and-
research/2014_strategy_fact_sheet.pdf
Patterson, G. R., DeBaryshe, B. D., & Ramsey, E. (1989). A developmental perspective on
antisocial behavior. American Psychologist, 44, 329–335.
Pilgrim, C. C., Schulenberg, J. E., O'Malley, P.,M., Bachman, J. G., & Johnston, L. D. (2006).
Mediators and moderators of parental involvement on substance use: A national study of
adolescents. Prevention Science, 7(1), 75-89.
Pitel, L., Madarasova Geckova, A., Kolarcik, P., Halama, P., Reijneveld, S.A., & van Dijk, J.P.
(2012). Gender differences in the relationship between religiosity and health-related
behaviour among adolescents. Journal of Epidemiology and Community Health, 66(12),
1122-1128.
Pleck, J.H., & Masciadrelli, B.P. (2004). Paternal involvement by U.S. residential fathers:
Levels, sources and consequences. In M.E. Lamb (Ed.), The role of the father in child
development (4th ed., pp. 222-271). Hoboken, NJ: Wiley.
Popovici, I., Homer, J.F., Fang, H., and French, M.T. (2012). Alcohol use and crime: findings
from a longitudinal sample of U.S. adolescents and young adults. Alcoholism, Clinical and
Experimental Research, 36(3), 532- 543.
Prohibition of Alcohol. (2009). In P. Korsmeyer & H. R. Kranzler (Eds.), Encyclopedia of
Drugs, Alcohol & Addictive Behavior (3rd ed., Vol. 3, pp. 303-307). Detroit: Macmillan
Reference USA.
Ramirez, R., Hinman, A., Sterling, S., Weisner, C., & Campbell, C. (2012). Peer influences on
adolescent alcohol and other drug use outcomes. Journal of Nursing Scholarship, 44(1),
36-44
Randall, D., and Wong, W.R. (1976). Drug education to date: A review. Journal of Drug
Education, 6, 1-21.
Reifman, A., Barnes, G. M., Dintcheff, B. A., Farrell, M. P., & Uhteg, L. (1998). Parental and
peer influences on the onset of heavier drinking among adolescents. Journal of Studies on
Alcohol, 59, 311–317.
79
Regnerus, M.D. (2003). “Moral communities and adolescent delinquency: Religious contexts
and community social control.” Sociological Quarterly, 44(4), 523-554.
Rogosa, D. R. (1987). Causal models do not support scientific conclusions: A comment in
support of Freedman. Journal of Educational Statistics, 12, 185-195.
Rote, S. M., & Starks, B. (2010). Racial/Ethnic differences in religiosity and drug use. Journal of
Drug Issues, 40(4), 729-753.
Salas-Wright, C., Vaughn, M., Hodge, D., & Perron, B. (2012). Religiosity Profiles of American
Youth in Relation to Substance Use, Violence, and Delinquency. Journal of Youth and
Adolescence, 41(12), 1560-1575.
Sawyer, T. M., & Stevenson, J. F. (2008). Perceived parental and peer disapproval toward
substances: Influences on adolescent decision-making. Journal of Primary Prevention,
29(6), 465-77.
Sawyer, T., & Stevenson, J. (2008). Perceived Parental and Peer Disapproval Toward
Substances: Influences on Adolescent Decision-Making. The Journal of Primary
Prevention, 29(6), 465-477.
Scholte, R. H. J., van Lieshout, C. F. M., & van Aken, M. A. (2001). Perceived relational support
in adolescence: Dimensions, configurations, and adolescent adjustment. Journal of
Research in Adolescence, 11, 71–94.
Sloane, D., and Potvin, R.H. (1986). “Religion and delinquency: Cutting through the Maze.”
Social Forces, 65, 87–105.
Stark, R. (1996). “Religion as Context: Hellfire and Delinquency One More Time.” Sociology of
Religiosity, 57, 163–73.
Steinberg, L. and Fletcher, A. (1994). Parental monitoring and peer influences on adolescent
substance use. Pediatrics, 93(6), 1060-1064.
Substance Abuse and Mental Health Services Administration. (2013). Results from the 2013
National Survey on Drug Use and Health: Summary of National Findings, NSDUH Series
H-44, HHS Publication No. (SMA) 12-4713. Rockville, MD: Substance Abuse and Mental
Health Services Administration. Retrieved from
http://www.samhsa.gov/data/sites/default/files/NSDUHresultsPDFWHTML2013/Web/NS
DUHresults2013.pdf
Svensson, R. (2003). Gender Differences in Adolescent Drug Use: The Impact of Parental
Monitoring and Peer Deviance. Youth & Society, 34(30, 300-329.
Tanner-Smith, E.E. (2012). Pubertal Development and Adolescent Girls’ Substance Use: Race,
Ethnicity, and Neighborhood Contexts of Vulnerability. The Journal of Early Adolescence,
32(5), 621-649.
Thai, N.D., Connell, C.M., and Tebes, J.K. (2010). Substance Use Among Asian American
Adolescents: Influence of Race, Ethnicity, and Acculturation in the Context of Key Risk
and Protective Factors. Asian American Journal of Psychology. 1(4), 261-274.
The U.S. Department of Health and Human Services. (2007). The surgeon general's call to action
to prevent and reduce underage drinking. Rockville, MD.
Tittle, C., & Welch, M. (1983). Religiosity and deviance: Toward a contingency theory of
constraining effects. Social Forces, 61, 653-682.
Trim, R.S., & Chassin, L. (2008). Neighborhood Socioeconomic Status Effects on Adolescent
Alcohol Outcomes Using Growth Models: Exploring the Role of Parental Alcoholism.
Journal of Studies on Alcohol and Drugs, 69(5), 639–648.
80
Trucco, E.M., Colder, C.R., & Wieczorek, W.F. (2011). Vulnerability to peer influence: A
moderated mediation study of early adolescent alcohol use initiation. Addictive
Behaviors, 36(7), 729-736.
Tucker, J.S., de la Haye, K., Kennedy, D.P., Green Jr, H.D., & Pollard, M.S. (2014). Peer
Influence on Marijuana Use in Different Types of Friendships. Journal of Adolescent
Health, 54(1), 67-73.
Van der Vorst, H., Engels, R. C. M. E., Meeus, W., & Dekovic, M. (2006). Parental attachment,
parental control, and early development of alcohol use: A longitudinal study. Psychology
of Addictive Behaviors, 20, 107–116.
Vaughan, E. L., de Dios, M. A., Steinfeldt, J. A., & Kratz, L. M. (2011). Religiosity, alcohol use
attitudes, and alcohol use in a national sample of adolescents. Psychology of Addictive
Behaviors, 25(3), 547-553.
Wallace, J. M. J., Forman, T. A., Caldwell, C. H., and Willis, D. S. (2003). Religion and U.S.
secondary school students: Current patterns, recent trends, and socio-demographic
correlates. Youth and Society, 35(1), 98–125.
Wallace, J. M., Jr., Brown, T. N., Bachman, J. G., & LaVeist, T. A. (2003). The influence of race
and religion on abstinence from alcohol, cigarettes and marijuana among adolescents.
Journal of Studies on Alcohol, 64, 843-848.
Wallace, J. M., Jr., Yamaguchi, R., Bachman, J. G., O'Malley, P. M., Schulenberg, J. E., &
Johnston, L. D. (2007). Religiosity and adolescent substance use: The role of individual
and contextual influences. Social Problems, 54(2), 308-327.
Wallace, J.M., Delva, J., O'Malley, P.M., Bachman, J.G., Schulenberg, E., Johnston, L.D., and
Stewart, C. (2007). Race/Ethnicity, Religiosity and Adolescent Alcohol, Cigarette and
Marijuana Use. Social Work in Public Health. 23(2-3), 193-213
White, W.L. (1998). Slaying the dragon: The history of addiction treatment and recovery in
America. Bloomington, IL: Chestnut Health Systems.
Whitney, S. D., Kelly, J. F., Myers, M. G., and Brown, S. A. (2002). "Prenatal Substance Use,
Family Support and Outcome Following Treatment for Adolescent Psychoactive
Substance Use Disorders." Journal of Child & Adolescent Substance Abuse, 11(4),67-80.
Wills, T. A., & Cleary, S. D. (1999). Peer and adolescent substance use among 6th–9th graders:
Latent growth analyses of influence versus selection mechanisms. Health Psychology,
18(5), 453-463.
Wills, T. A., and Cleary, S. D. (1996). "How are social support effects mediated? A test with
parental support and adolescent substance use". Journal of personality and social
psychology, 71(5), 937-952.
Wills, T. A., Resko, J. A., Ainette, M. G., & Mendoza, D. (2004). Role of parent support and
peer support in adolescent substance use: A test of mediated effects. Psychology of
Addictive Behaviors, 18(2), 122-134.
Wills, T. A., Yaeger, A. M., & Sandy, J. M. (2003). Buffering effect of religiosity for adolescent
substance use. Psychology of Addictive Behaviors, 17(1), 24-31.
Windle, M. (1994). A study of friendship characteristics and problem behaviors among middle
adolescents. Child Development, 65, 1764–1777.
Windle, M. (2000). Parental, sibling, and peer influences on adolescent substance use and
alcohol problems. Applied Developmental Science, 4, 98–110.
Windle, M. (2000). Parental, sibling, and peer influences on adolescent substance use and
alcohol problems. Applied Developmental Science, 4, 98–110.
81
Wodarski, J. S. (2010). Prevention of adolescent reoccurring violence and alcohol abuse: A
multiple site evaluation. Journal of Evidence-Based Social Work, 7(4), 280-301.
Youniss, J., & Smoller, J. (1985). Adolescent relations with mothers, father, and friends.
Chicago: University of Chicago Press.